JP3570138B2 - 3D stereolithography data creation method - Google Patents

Description

【００６８】
そして１番目のデータから順次ループ番号、レベル、回転方向のデータを読み込み、その回転方向を作業領域ＷＯＲＫ（レベル０）に格納する。そして当該データのレベルが０であるのか否かを判断する。ここで１番目のデータであるループＬＰ１のレベルは０であるから、親ループと判断され、次の２番目のデータを読み出し、その回転方向を作業領域ＷＯＲＫ（レベル１）に書き込む。
【００６９】
ここで２番目のデータであるループＬＰ２はレベル１であるから、次に回転方向がレベル０の親ループＬＰ１と等しいか判断される。
ここでは異なるから、次に３番目のループＬＰ４のデータを読み出し、その回転方向を作業領域ＷＯＲＫ（レベル２）に書き込み込む。この場合レベルは２であるので、回転方向を、レベル１のループの回転方向と比較する。ここで２番目のループＬＰ２の回転方向とループＬＰ４の回転方向が異なるので、次の４番目のループＬＰ５のデータを読み出し、そのレベル３に対応する作業領域ＷＯＲＫ（レベル３）に回転方向を書き込む。ここでレベル３であるから先のレベル２のループＬＰ４の回転方向と比較する。この場合回転方向が異なるので、５番目のデータであるループＬＰ３のデータを読み出し、そのレベル１に対応する作業領域ＷＯＲＫ（レベル１）に回転方向を書き込む。ここでレベルが１であるので、レベル０のループＬＰ１の回転方向と比較する。この場合両ループＬＰ１、ＬＰ３とも回転方向が同じであるから、５番目のループ番号のデータを削除する。
そして現在呼び出しているループのデータとして親ループのレベル、つまり元をセットし、次に６番目のデータ、つまりループＬＰ６のデータを読み出し、もしこれで全てのデータの読み出しが終了していなければ、次の７番目のデータの読み出しに移るが、図３１（ａ）の場合、終了となるので、ループＬＰ６のレベルと元レベルとを比較し元レベルが等しく若しくは大きい場合にはループＬＰ６のレベル、つまり２から１減じてループＬＰのレベルを１にセットする。
【００７０】
このようにして、不具合のあるデータであってもループの回転方向により不要なループを図３１（ｂ）に示すように判定して削除でき、これに起因する形状不具合の防止や、データ修正工数の削減が図れる。
さて不要な閉ループの削除過程が終了すると、図１０に示すように閉ループの回転方向の是正の過程に移行する。この閉ループの回転方向の是正では、図３２に示すフローチャートによる処理が為される。この場合上述と同様に一断面内の全ての閉ループの包含関係を調査してメモリにループ番号とレベルとをセットする。ここで例として図３１（ａ）の場合を取り上げる。
【００７１】
このセットされたデータからまず１番目のデータを順次読み出し、読み出してデータのレベルが奇数か、偶数かを見て、奇数のレベルのループ番号のデータを回転方向ＣＷに変換し、偶数の場合には回転方向ＣＣＷに変換する。
このようにすることによりループの回転方向に信頼性が無くとも、断面内のループの包含関係により内外を判断し、それに応じて回転方向を是正することにより、これに起因する形成不具合の防止、データ修正工数の削減が図れる。
【００７２】
以上のようにして図１０に示す不具合補正の処理を終了して不具合補正後の断面輪郭線データをデータ生成装置３２は作成し、図１に示すように形成寸法精度向上のために断面輪郭線データを補正する段階に入る。
この段階は、図３３に示すように形成後の表面処理の方法に応じたデータ補正処理と、光硬化性樹脂の余剰硬化を是正するためのデータ補正処理とがあり、形成後の表面処理の方法に応じたデータ補正処理の場合には、形成後に磨きによる表面処理を行う場合の補正方法と、形成後にコーティングによる表面処理を行なう場合の補正方法とがある。
【００７３】
前者の補正方法は図３４に示すフローチャートによる処理が為される。
つまり造成物の総積層数をＮ_ｍａｘ とし、第ｎ層と、その前の層である第ｎ−１層の論理和を演算して第ｎ層の補正データとする処理を行うのである。但し、第０層のデータは三次元形状の最低Ｚ値でスライスした断面輪郭線データとする。従って補正前のデータが図３５（ａ）の場合、補正後のデータは図３５（ｂ）に示すように上向き部分、下向き部分とも所望の形状の外側に位置した形状データとなる。この補正後の形状データにより造形物を形成した後、外側の部分を研磨処理するだけで、所望の寸法が確保でき、容易な作業で正確な寸法の造形物１が図３５（ｃ）のように得られることになる。
【００７４】
以上のような補正したデータを用いて造形物を形成した後に形成段差を研磨処理するだけで、所望の寸法が確保でき、容易な作業で正確な寸法の造形物１を得ることができる。
後者の補正方法は、図３６に示すフローチャートによる処理が為される。
つまり造成物の総積層数をＮ_ｍａｘ とし、第ｎ層と、その前の層である第ｎ−１層の論理積を演算して第ｎ層の補正データとする処理を行うのである。但し、第０層のデータは三次元形状の最低Ｚ値でスライスした断面輪郭線データとする。
【００７５】
従って補正前のデータが図３７（ａ）の場合、補正後のデータは図３７（ｂ）に示すように所望の形状の外側に位置していた下向き部分も内側に位置した形状データとなる。この補正後の形状データにより造形物を形成した後、段差部分ををコーティング処理により埋めることで、所望の寸法が確保でき、容易な作業で正確な寸法の形状の造形物１が図３７（ｃ）のように得られることになる。
【００７６】
以上のように研磨若しくはコーティング処理に対応したデータ補正を行った後、光硬化性樹脂の余剰硬化を是正するためのデータ補正処理を行う。
この光硬化性樹脂の余剰硬化を是正するためのデータ補正処理としては、次のような方法がある。
まずその一つは、実験或いは計算機解析により、条件に応じた寸法偏差表をデータベースとして予め登録しておく。表２はこのデータベースの一例を示し、この場合第１層から第五層迄の寸法偏差を示しており、第五層以上は同一の値とする。
【００７７】
【表２】

【００７８】
この処理は図３８に示すフローチャートにより為される。
つまり最大積層数に対応した層Ｎ_ｍａｘ から順次下層へ向けて補正処理を行うのである。つまり第ｎ層の寸法偏差を条件とデータベースの数値により算出し、寸法偏差分下方の層の層番号を求める。但し算出した数値で示される位置がある層の中間を示す場合にはその層の一つ上位の層の層番号を求める。この層番号ｍが０層より大きくなければ、第ｎ層の補正データは空データ（データ無し）とし、また層番号ｍが０層より大きければ、第ｎ層と第ｍ層の論理積演算を行い、第ｎ層の補正データとするのである。
【００７９】
この補正処理を図３９の形状データに基づいて詳説する。図３９の形状データが最大積層数が１００層であり、１００乃至９６層の各層の厚みが０．２ｍｍ、９５層乃至８６層の各層の厚みが０．２ｍｍとする。
（１）最上位の１００層目は最上位層であるため補正は行わない。
（２）９９層目 ０．２ｍｍの２層で、寸法偏差が０．２５ｍｍ、０．２５ｍｍ下方は、９８層目
（３）９８層目 ０．２ｍｍの３層で、寸法偏差が０．３ｍｍ、０．３ｍｍ下方は、９７層目
（４）９７層目 ０．２ｍｍの４層で、寸法偏差が０．３３ｍｍ、０．３３ｍｍ下方は、９５層目
（５）９６層目 ０．２ｍｍの５層で、寸法偏差が０．３５ｍｍ、０．３５ｍｍ下方は、９３層目
（６）９５層目 ０．１ｍｍの１層と０．２ｍｍの５層で、寸法偏差が０．２５ｍｍ＋（０．３５−０．１５）＝０．４５ｍｍで、０．４５ｍｍ下方は、９１層目
（７）９４層目 ０．１ｍｍの２層と０．２ｍｍの５層で、寸法偏差が０．４ｍｍ＋（０．３５−０．２５）＝０．５ｍｍで、０．５ｍｍ下方は、８９層目
（８）９３層目 ０．１ｍｍの３層と０．２ｍｍの５層で、寸法偏差が０．５ｍｍ＋（０．３５−０．３）＝０．５５ｍｍで、０．５５ｍｍ下方は、８８層目
（９）９２層目 ０．１ｍｍの４層と０．２ｍｍの５層で、寸法偏差が０．５３ｍｍ＋（０．３５−０．３３）＝０．５５ｍｍで、０．５５ｍｍ下方は、８７層目
（１０）９１層目 ０．１ｍｍの５層で、寸法偏差が０．５５ｍｍで、０．５５ｍｍ下方は、８６層目
以下の層も同様にして、補正対象の層と求めた下方の層の２層の論理積（重なり部分）演算して補正対象層の補正データとするのである。
【００８０】
データ補正を行って造形物形成を行った場合の流れを示しており、図４０（ａ）に示す形成データから上記の処理により図４０（ｂ）に示す補正したデータを得、この補正後データにより造形物１を形成した場合余剰硬化部分Ｘがあっても所望する形状に近い形状の造成物１が得られることが分かる。
つまり従来では不可能な一層の厚みが部分によって異なる場合であっても光硬化性樹脂の余剰硬化の補正が実現できる。また従来補正を実施することによりかえって不具合となる、最上部が杯状に開いた形状であっても、正確な形状に補正することが可能となる。更に或る層の補正を行う場合に、論理積の演算は１回のみ行えば良いので、処理時間が短縮する。
【００８１】
図３８に示す処理では、図４１（ａ）乃至（ｄ）に示すような形状の場合、下向き面の最上先端部分に形状欠落部分が図４１（ｄ）に示すように発生し補正により寸法精度が低下する恐れがある。尚図４１（ａ）は元の形状データ、（ｂ）は不具合がある補正前の形状データ、（ｃ）は補正後のデータ、（ｄ）はこの補正後のデータにより造形物１を形成した状態を示す。
【００８２】
この点を補うために、補正を行わない層範囲、補正を行う層範囲のまとまりを夫々複数指定可能とするのが図４２に示す処理方法である。つまり補正対象範囲を指定入力し、その処理しようとする層が対象範囲である層なのか否かを判定し、否であれば補正をキャンセルするのである。
これを図４３（ａ）乃至（ｄ）に示す具体例により説明すると、図４３（ａ）の元のデータから得られた不具合を含むデータ（図４３（ｂ））において、補正を行なわない範囲と補正を行う範囲を指定して、データ補正を行い、図４３（ｃ）に示すような形状データを作成し、この補正後の形状データに基づいて造形物１を図４３（ｄ）に示すように形成すると、欠落部分が無くなり、段差のある部分であっても、補正に伴う形状の不具合を抑制し、正確な形状の造形物１を得ることができる。
【００８３】
図４４（ａ）乃至図４４（ｄ）は、別の例を示しており、図４４（ａ）の元のデータから得られた不具合を含むデータ（図４４（ｂ））から補正を行う範囲を二つに分け、夫々に於いて補正を行い、補正後のデータ（図４４（ｃ））により造形物１を図４４（ｄ）に示すように形成すると、上記の場合と同様に欠落部分が無くなり、段差のある部分であっても、補正に伴う形状の不具合を抑制し、正確な形状の造形物１を得ることができる。
【００８４】
上記のように補正を行う層の範囲、行わない層の範囲の纏まりを適切に支持することによって、段差のある形状であっても、補正に伴う形状の不具合を抑制し、正確な形状の造形物１を得ることができる。
ところで、図３８又は図３９に示す余剰硬化の補正処理は、比較的容易な手順・演算により補正を行うことが可能であるが、あらゆる形状に対して正確な補正を保証するものではない。
【００８５】
つまりこれら処理方法においては、ある層に対する補正が、当該層全域に渡って同一の条件で行われるからである。実際の形状では、一つの層においても部分部分でその上に重なる層の条件が異なるために、余剰硬化の寸法偏差が異なってくる場合がある。図４５のような形状の正しい補正を行うことができない。
つまり図４５（ａ）に示すような形状の元のデータを、図４５（ｂ）に示す不具合を持つ補正前のデータに変換し、更に図３８の処理方法によりデータを補正して図４５（ｃ）に示すようなデータを作成し、このデータにより造営物１を形成すると、図４５（ｄ）に示すように欠落部分が発生する。一方、図４２に示す処理方法により図４５（ｅ）に示すように補正範囲を指定してデータ補正を行い、この補正後データで造形物１を図４５（ｆ）に示すように形成した場合には形状過多な部分が発生する。
【００８６】
このような点を考慮してあらゆる形状に対して余剰硬化の補正を行う方法が次に述べる処理である。
つまり本方法では、予めレーザー走査条件毎に、一層の厚み及び積層数によって異なる寸法偏差を実験或いは計算機解析により求めて予めデータベース化しておく。
【００８７】
そして図４６に示すフローチャートの処理を実行するのである。この処理では図に示すように各層の補正を行う際に、補正対象となる層を、その層の上位に存在する複数層の形状によって重なり合う層数の条件毎に分割し、夫々の部分における寸法偏差を当該部分の条件に応じてデータベース（表２と同じ）の数値から算出し、次にその補正値に相当する分、下方の層を求め、更にその層と補正対象となる層の当該部分形状との重なり合う形状を各々求め、それらの形状と統合して補正対象となる層の補正データとするのである。
【００８８】
これを全層に亘って行うことにより求める補正データを得る。但し、最上位となる部分については、補正を行わない。
この補正処理を更に具体的な形状について説明する。
図４７に示す、最上位層が１００層目、段差部分の層が８９層であるような形状データを補正するに際しては、
（１）最上位の１００層目は最上位層であるため補正は行わない。
【００８９】
（２）９９層目 ０．１ｍｍの２層で、寸法偏差が０．４ｍｍ、０．４ｍｍ下方は、９５層目で、論理積をとる。
（３）９８層目 ０．１ｍｍの３層で、寸法偏差が０．５ｍｍ、０．５ｍｍ下方は、９３層目で、論理積をとる。
（４）９７層目 ０．１ｍｍの４層で、寸法偏差が０．５３ｍｍ、０．５３ｍｍ下方は、９２層目で、論理積をとる。
【００９０】
（５）９６層目 ０．１ｍｍの５層で、寸法偏差が０．５５ｍｍ、０．５５ｍｍ下方は、９１層目で、論理積をとる。
（６）９５層目乃至９０層は同様にして９０層乃至８５層目と論理積をとる。
これら１００層乃至９０層までは層善意に亘って条件が同一になるため分割されず、図３８の処理方法と同様に補正対象の層と求めた下方の層の二層の論理積（重なり部分）を演算して補正対象層の補正データとする。
【００９１】
（７）８９層目は図４８（ａ）に示すように部分▲１▼は最上位であるが、部分▲２▼は０．１ｍｍの５層で、寸法偏差０．５５ｍｍとなり、８４層目と論理積を取る。図４８（ｂ）は８９層目の補正結果を示す。
（８）８８層目は図４９（ａ）に示すように部分▲１▼は０．１ｍｍの２層で、寸法偏差０．４ｍｍとなり、８４層目と論理積を取る。また部分▲２▼は０．１ｍｍ５層で、寸法偏差０．５５ｍｍとなり８３層目と論理積をとる。図４９（ｂ）は８８層目の補正結果を示す。
【００９２】
以下同様にして補正を行う。
図５０（ａ）は補正後の形状データを示し、この形状データを用いて造営物１の造形を行った結果が図５０（ｂ）であり、この結果形状の欠落部分や過多な部分がなくなり、正確な形状になる。
尚上記説明で用いた形状は説明のため単純化した形状としているが、任意の形状であっても同様な考え方で処理が可能である。
【００９３】
このように図４６に示す処理方法によれば、部分毎に余剰硬化による寸法偏差の補正が行えるため、如何なる形状であっても正確な補正が実現できる。
尚各層の形状分割は次のような方法で行う。
例えば今上から見た形状が図５１（ａ）に示すような形状で側面から見た形状が図５１（ｂ）に示すような形状の場合、図５２（ａ）に示すように補正対象層Ｌｔと、一層上位層Ｌ１との差の内、上位層にない部分Ａ１を分割する。
【００９４】
つまりＡ１（最上位層として処理する部分）は層Ｌ１が重ならない保証対象層Ｌｔの部分で最上位層として処理される。図中Ｂ１は層Ｌ１が重なる補正対象層Ｌｔの残りの部分を示す。
次に図５２（ｂ）に示すように残りの部分Ｂ１と、二層上位層Ｌ２との差の内、上位層にない部分Ａ２を分割する。図中Ｂ２は層Ｌ２に重なった補正対象層Ｌｔの残りの部分を示す。
【００９５】
以下同様に残りの部分とその更に上位層との差の内、上位層の重ならない部分を切り取る操作を繰り返すことにより分割を行う。
この場合ある程度以上は寸法偏差が一定値になるので、上位３〜４層程度まで行えば十分である。
図５３（ａ）は残りの部分Ｂ２と、三層上位層Ｌ３との差の内、層Ｌ３が重ならない補正対象層Ｌｔの部分Ａ３を分割する状態を示す。図中Ｂ３は層Ｌ３に重なった補正対象層Ｌｔの残りの部分を示す。
【００９６】
図５３（ｂ）は残りの部分Ｂ３と、三層上位層Ｌ４との差の内、層Ｌ４が重ならない補正対象層Ｌｔの部分Ａ４を分割する状態を示す。図中Ｂ４は層Ｌ４に重なった補正対象層Ｌｔの残りの部分を示す。
上記の分割した結果は図５４に示すようになり、これらの部分について夫々の条件に応じて補正を行えば良い。
【００９７】
次にこの例における各部分形状について、Ａ１について補正なし、Ａ２について一層下との論理積、Ａ３について二層下との論理積、Ａ４について三層下との論理積、Ａ５について四層下との論理積というように、補正したところ、図５５（ａ）に示すような部分形状毎の補正結果が得られ、これら部分形状結果を統合すると図５５（ｂ）に示すようになる。尚図５５（ｂ）の（イ）に示す部分が補正後の層形状を、補正前の層形状は長方形全体（ロ）を示す。
【００９８】
【発明の効果】
請求項１の発明は、光硬化性樹脂に光を照射して形成した光硬化層、或いは薄板を複数層積み重ねて所望の三次元形状を形成する三次元形状装置に用いる三次元光造形用データ作成方法において、三次元形状データを形成段差が目立たないようにスライスして複数の断面輪郭線データを作成する段階と、入力された形状データの不備による断面輪郭線データの不具合を補正する段階と、形成寸法精度向上のために断面輪郭線データを補正する段階とから成るので、三次元形状形成用のデータ作成工数と造形物の後加工処理を軽減し、三次元形状形成作業のコスト低減、納期短縮、造形物の高精度化を実現することができるという効果がある。
【００９９】
また、スライスにより断面輪郭線データを作成する段階は、不具合の出る座標値で三次元形状データをスライスするのを避けるための座標値補正過程と、形成段差を目立たなくするために、形成段差の目立つ箇所を通常ピッチより細かくスライスする過程とから成るので、段差の目立たない造形物を得ることができ、そのため造形物の後加工処理を軽減できるという効果がある。
【０１００】
請求項２の発明は、請求項１の発明において、上記スライス座標値補正過程は、条件により計算或いは指定されたスライス座標値列の各々の数値に対して、データ作成対象となる三次元形状形成装置の積層方向移動量の最小値単位未満の任意の微小数値を一律に加算或いは減算することにより行うので、形状データをスライスしてデータ補正を行う際に、不具合のでる位置でのスライスを防ぎ、これに起因する形成不具合の防止、データ修正工数の削減が図れるという効果がある。
【０１０１】
請求項３の発明は、請求項１の発明において、上記スライスする過程は、形状を一定ピッチでスライスするとともに、該スライスにより形成したある二つの層間を補間する層が必要か否かを、この二つの層で挟まれた領域に存在するＳＴＬデータのファセットの法線ベクトル情報から判断して補間層数を決定し、その数値に基づいてスライス位置を決定して２層間を追加スライスするので、細かいピッチでのスライスが必要な箇所を自動的に認識してスライスデータを作成することが可能となり、そのため見落としやテスト形成が必要なくなり、作業工数を増加させることなく段差の目立たない造形物の形成が可能となるという効果がある。
【０１０２】
請求項４の発明は、請求項３の発明において、上記二つの層で挟まれた領域に存在するＳＴＬデータのファセットの法線ベクトルの内、下向きの法線ベクトルを持つデータを判断から外すので、もともと段差が目立たない下向きの面について補正しないため、その分積層数の削減が図れ、データ処理時間、形成時間を短縮することが可能となり、しかも積層厚みを細かくしないため、光硬化性樹脂の余剰硬化によって起きる寸法偏差の増大からの寸法精度の低下を抑制できるという効果がある。
請求項５の発明は、光硬化性樹脂に光を照射して形成した光硬化層、或いは薄板を複数層積み重ねて所望の三次元形状を形成する三次元形状装置に用いる三次元光造形用データ作成方法において、三次元形状データを形成段差が目立たないようにスライスして複数の断面輪郭線データを作成する段階と、入力された形状データの不備による断面輪郭線データの不具合を補正する段階と、形成寸法精度向上のために断面輪郭線データを補正する段階とから成り、上記断面輪郭線データの不具合を補正する段階は、開ループを閉ループ化する過程と、閉ループの自己交差を分割・削除する過程と、閉ループの重なりを是正する過程と、不要な閉ループを削除する過程と、閉ループの回転方向が正しいかどうが不明確なデータに対して閉ループの回転方向の予め規定した側が形状の内側になるように閉ループの回転方向を是正する過程とから成るので、不具合を含んだ断面輪郭線データであっても、自動的に判断して問題のない断面輪郭線データに変換するので、三次元形状形成作業のコスト低減、納期短縮、造形物の高精度化を実現することができる上に、データ修正工数を大幅に削減できるという効果がある。
【０１０３】
請求項６の発明は、請求項５の発明において、上記開ループを閉ループ化する段階は、開ループが自己で閉ループ化を試行し、閉ループ化されない場合は、他の開ループとの接続による閉ループ化を試行し、他の開ループと接続できない場合は削除するので、断面輪郭線データが開ループとなっている不具合を閉ループ化して不具合を無くし、且つ不具合が解消しない断面輪郭線データを削除してその影響を除去することができるという効果がある。
【０１０４】
請求項７の発明は、請求項６の発明において、上記開ループの自己での閉ループ化の方法は、開ループに自己交差がある場合又は端点が当該開ループ上にある場合、交点で閉ループ化して残りを削除し、次に端点間の距離が指定値以下であれば、端点間に直線を追加して閉ループ化し、更に該閉ループと端点との距離が指定値以下であれば、当該端点から閉ループの垂線を追加して閉ループ化し、残りを削除するので、複数の開ループの閉ループ化に対応できるという効果がある。
【０１０５】
請求項８の発明は、請求項６又は７の発明において、上記複数の開ループを接続しての閉ループ化の方法は、ループの方向が既に正しく設定されているとする条件で、或る二つの開ループが交差或いは接点を持つ場合に対して、ループの方向が合うように交点で二つの開ループをつなぎ替え、短い方を削除し、残った開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
終点と他の開ループの始点との距離、或いは始点と他の開ループの終点との距離が指定値以下の場合に対して、複数の該当があれば最短距離のループを対象として端点間に直線を追加することにより接続し、この接続によりできた開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
他の開ループとの間で端点とループとの距離が指定値以下であるという関係の場合に対して、当該端点からループへの垂線を追加し、該垂線とループとの交点からループ方向が合ように接続するとともに接続によって残った部分を削除し、該接続と削除によりできた開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
上記各過程において処理対象となる開ループ以外の開ループのデータを削除する過程とからなり、元から存在している開ループ及び上記過程で二次的に発生した開ループのデータが無くなるまで、上記過程の処理を繰り返すので、ループの方向が既に正しく設定されているとする条件で、複数の開ループの閉ループ化に対応できるという効果がある。
【０１０６】
請求項９の発明は、請求項６又７の発明において、複数の開ループを接続しての閉ループ化の方法は、ループの方向が不明確な条件で、或る二つの開ループが交差或いは接点を持つ場合に対して、各々のループの、交点若しくは接点で分割された長い方同士を接続するとともにループの方向を二つの内の長い方に合わせ且つ残りを削除し、得られた開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
他の開ループとの間で端点とループとの距離が指定値以下の場合に対して、複数の該当があれば最短距離のループを対象として端点間に直線を追加することにより接続するとともにループの方向を二つの内の長い方に合わせ、得られた開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
他の開ループとの間で端点とループとの距離が指定値以下であるという関係の場合に対して、当該端点からループへの垂線を追加し、該垂線とループとの交点で分割された長い方と接続するとともにループの方向を二つの内の長い方に合わせ且つ残りを削除し、得られた開ループについて請求項７の開ループの自己での閉ループ化の方法により閉ループ化を試行し、該試行により閉ループ化されれば該当する両開ループについての処理を終了したとして、次の開ループの処理に移行し、試行で閉ループ化されなければ、当該開ループのデータを処理データの末尾に追加する過程と、
上記各過程において処理対象となる開ループ以外の開ループのデータを削除する過程とからなり、元から存在している開ループ及び上記過程で二次的に発生した開ループのデータが無くなるまで、上記過程の処理を繰り返すので、ループの方向が不明確であっても、複数の開ループの閉ループ化に対応できるという効果があるという効果がある。
【０１０７】
請求項１０の発明は、請求項５の発明において、上記自己交差を分割・削除する過程は、自己交差の交点毎に閉ループの切り出しと、残りのループの接続による閉ループ化を行っていくことにより最終的に複数の閉ループを得、該面積最大の閉ループのみ以外を削除するので、閉ループの自己交差の分割、削除が自動的にできるという効果がある。
【０１０８】
請求項１１の発明は、請求項５の発明において、上記閉ループの重なりを是正する過程は、ループの回転方向が既に正しく設定されている場合において、重なった二つの閉ループの回転方向が等しい時に、当該両閉ループの論理和を取って両閉ループを統合し、回転方向が異なる時に、面積大の閉ループから面積小の閉ループを引く形で両閉ループを統合するので、ループの回転方向が正しく設定されている条件で、閉ループの重なりの是正が自動的にできるという効果がある。
【０１０９】
請求項１２の発明は、請求項５の発明において、上記不要な閉ループを削除する過程は、ループの回転方向が既に正しく設定されている場合に、一断面内の全ての閉ループの包含関係を予め求め、この求めた包含関係情報を用いて最上位の閉ループを親ループとし、該親ループと該親ループの一つ下の閉ループの回転方向の相違をチェックして親ループと同じ回転方向を持つ場合には該閉ループを削除し、削除した閉ループに下位の閉ループが包含される場合には包含される閉ループを一段上位に上げ、該閉ループについても親ループとの間で回転方向の相違をチェックし、親ループと同じ回転方向を持つ場合には該閉ループを削除するという操作を親ループから最下位の閉ループまで順番に行うので、ループの回転方向が正しく設定されている条件で、不要な閉ループを自動的に削除することができるという効果がある。
【０１１０】
請求項１３の発明は、請求項５の発明において、上記閉ループの回転方向を是正する過程は、一断面内の全ての閉ループの包含関係を求め、この求めた包含関係情報を用いて最上位の閉ループを親ループとして該親ループに対して内側が形状の内側になるような回転方向を与え、以下最下位の閉ループまで順番に夫々親ループと反対のループ回転方向を設定するので、ループの回転方向に信頼性が無くとも、断面内のループの包含関係により内・外を判断し、それに応じて回転方向を是正することで、これに起因する形成不具合の防止が図れ、またデータ修正工数の削減が図れるという効果がある。
【０１１１】
請求項１４の発明は、光硬化性樹脂に光を照射して形成した光硬化層、或いは薄板を複数層積み重ねて所望の三次元形状を形成する三次元形状装置に用いる三次元光造形用データ作成方法において、三次元形状データを形成段差が目立たないようにスライスして複数の断面輪郭線データを作成する段階と、入力された形状データの不備による断面輪郭線データの不具合を補正する段階と、形成寸法精度向上のために断面輪郭線データを補正する段階とから成り、形成寸法精度向上のために断面輪郭線データを補正する段階は、形成後の表面処理方法に応じたループの補正過程と、光硬化性樹脂の余剰硬化のためのデータ補正過程と成るので、三次元形状形成作業のコスト低減、納期短縮、造形物の高精度化を実現することができる上に、造形物の形成後に行う表面処理方法に対応して形成寸法精度の向上が図れるデータ補正ができるという効果がある。
【０１１２】
請求項１５の発明は、請求項１４の発明において、上記形成後の表面処理方法に応じたループの補正過程は、形成後に磨きによる表面処理を行う場合に対応して、或る層の断面輪郭線データと該層の一層下の層の断面輪郭線データとの論理和を該層の補正データとし、この補正データを生成する操作を全層に亘って行うので、形成後に形成段差を削り取ることによって、所望の寸法を確保でき、そのため容易な作業で正確な寸法の造形物を得ることができるという効果がある。
【０１１３】
請求項１６の発明は、請求項１４の発明において、上記形成後の表面処理方法に応じたループの補正過程は、形成後にコーティングによる表面処理を行う場合に対応して、或る層の断面輪郭線データと該層の一層下の層の断面輪郭線データとの論理積を該層の補正データとし、この補正データを生成する操作を全層に亘って行うので、形成後に形成段差を埋めるよに表面をコーティングすることによって、所望の寸法を確保でき、そのため容易な作業で正確な寸法の造形物を得ることができるという効果がある。
【０１１４】
請求項１７の発明は、請求項１４の発明において、上記光硬化性樹脂の余剰硬化のためのデータ補正過程は、条件によって異なる余剰硬化の寸法偏差を予め条件付けしてデータベースに登録しておき、補正の際に補正対象層における寸法偏差を、該補正対象層及びその上位に存在する層の厚みと層数の条件に応じてデータベースから検索若しくは検索した値を加工して得、この寸法偏差に相当する分下方に存在する層を求め、該層と上記補正対象層との断面輪郭線データの論理積を上記補正対象層の補正データとし、この補正データを得る操作を最上位層以外の全層に亘って行うので、従来技術では不可能な、一層の厚みが部分によってことなる場合であっても光硬化性樹脂の余剰硬化の補正が実現でき、従来技術では補正を実施することによりかえって不具合となる、最上位部が杯状に開いたような形状であっても、正確な形状に補正することが可能で、また或る層の補正を行う場合に、論理積の演算が一回のみ行えば良いので、処理時間が短縮するという効果がある。
【０１１５】
請求項１８の発明は、請求項１７の発明において、補正を行う層の範囲、補正を行わない層の範囲の纏まりを複数指定可能とし、補正を行う層範囲では一つの層範囲の纏まり毎に独立した形状データとして補正を行うので、段差のある形状であっても、補正に伴う形状の不具合を抑制し、正確な形状の造形物を得ることができるという効果がある。
【０１１６】
請求項１９の発明は、請求項１４の発明において、上記光硬化性樹脂の余剰硬化のためのデータ補正過程は、条件によって異なる余剰硬化の寸法偏差を予め条件付けしてデータベースに登録しておき、補正の際に補正対象層を当該層の上位に存在する複数層の形状によって層数の条件毎に複数の部分に分割し、上位に層が存在しない部分についてはその部分を補正データとし、その他の部分については夫々の部分における寸法偏差を、当該部分及び当該部分の上位に存在する層の厚みと総数の条件に応じてデータベースから検索若しくは検索した値を加工して得、この寸法偏差に相当する分下方に存在する層を求め、該層と上記当該部分との断面輪郭線データの論理積を上記当該部分の補正データとし、この補正データを分割された全ての部分について求めるとともにこれらの補正データを統合することにより補正対象層の補正データとし、この補正データを得る操作を全層に亘って行うので、あらゆる形状に対して余剰硬化の補正を行うことが可能なデータ補正ができるという効果がある。
【図面の簡単な説明】
【図１】本発明の実施形態の全体の処理の流れを示すフローチャートである。
【図２】同上の形成段差が目立たないように三次元形状データをスライスして複数の断面輪郭線を作成する段階のフローチャートである。
【図３】図２の段階におけるスライス座標値の補正過程を示すフローチャートである。
【図４】図２の補正過程を説明するための形状の断面図である。
【図５】図２の段階における段差の目立たないスイラス方法を示すフローチャートである。
【図６】図５のスライス方法を説明するための形状例図である。
【図７】図５におけるファセットの状態の説明図である。
【図８】図２は段階における段差の目立たないスライス方法の別の例を説明するための形状例図である。
【図９】図１に於ける断面輪郭線の不具合を補正する段階を説明するための不具合例図である。
【図１０】図１に於ける断面輪郭線の不具合を補正する段階のフローチャートである。
【図１１】図１０の段階における閉ループを閉ループ化する過程のフローチャートである。
【図１２】図１１における閉ループの自己での閉ループ化処理のフローチャートである。
【図１３】同上の閉ループ化処理の説明図である。
【図１４】同上の別の閉ループ化処理の説明図である。
【図１５】同上の他の閉ループ化処理の説明図である。
【図１６】同上のその他の閉ループ化処理の説明図である。
【図１７】図１１における複数の閉ループを接続しての閉ループ化処理のフローチャートである。
【図１８】同上の閉ループ化処理の説明図である。
【図１９】同上の別の閉ループ化処理の説明図である。
【図２０】同上の他の閉ループ化処理の説明図である。
【図２１】図１１における複数の閉ループを接続しての閉ループ化処理の別の例のフローチャートである。
【図２２】同上の閉ループ化処理の説明図である。
【図２３】同上の別の閉ループ化処理の説明図である。
【図２４】同上の他の閉ループ化処理の説明図である。
【図２５】図１０の段階における閉ループの自己交差を分割・削除する処理のフローチャートである。
【図２６】同上の閉ループの自己交差を分割・削除する処理の説明図である。
【図２７】図１０の段階における閉ループの重なり是正の処理のフローチャートである。
【図２８】同上の閉ループの重なり是正の説明図である。
【図２９】同上の閉ループ化の重なり是正の別の例の説明図である。
【図３０】図１０の段階における不要な閉ループを削除する処理のフローチャートである。
【図３１】同上の処理の説明図である。
【図３２】図１０の段階における閉ループの回転方向を是正する処理のフローチャートである。
【図３３】図１に於ける形成寸法精度向上のために断面輪郭線データの補正過程のフローチャートである。
【図３４】図３３における形成後の表面処理の方法に応じたデータ補正のフローチャートである。
【図３５】同上の処理の説明図である。
【図３６】図３３における形成後の表面処理の方法に応じたデータ補正の別の例のフローチャートである。
【図３７】同上の処理の説明図である。
【図３８】図３３における光硬化性樹脂の余剰硬化を是正するためのデータ補正のフローチャートである。
【図３９】同上の処理の説明図である。
【図４０】同上の処理の説明図である。
【図４１】図３８に示すデータ補正の問題点の説明図である。
【図４２】図３３における光硬化性樹脂の余剰硬化を是正するためのデータ補正の別の例のフローチャートである。
【図４３】同上の処理の説明図である。
【図４４】同上の処理の説明図である。
【図４５】図３８、図４２に示すデータ補正の問題点の説明図である。
【図４６】図３３における光硬化性樹脂の余剰硬化を是正するためのデータ補正の他の例のフローチャートである。
【図４７】同上の処理の説明図である。
【図４８】同上の処理の説明図である。
【図４９】同上の処理の説明図である。
【図５０】同上の処理の説明図である。
【図５１】同上の各層の形状分割方法の説明図である。
【図５２】同上の各層の形状分割方法の説明図である。
【図５３】同上の各層の形状分割方法の説明図である。
【図５４】同上の各層の形状分割方法の説明図である。
【図５５】同上の各層の形状分割方法の説明図である。
【図５６】従来例の説明図である。
【図５７】別の従来例の説明図である。
【図５８】他の従来例の説明図である。
【図５９】従来例の光硬化性樹脂の余剰硬化の補正での問題点の説明図である。
【図６０】従来例の光硬化性樹脂の余剰硬化の補正での問題点の説明図である。
【図６１】（ａ）は従来例のスライス座標値の補正での問題点の説明用の造形物の断面図である。
（ｂ）は従来例のスライス座標値の補正での問題点の説明用の造形物の斜視図である。
【図６２】図６１に示す断面輪郭線データの例図である。
【図６３】図６２の断面輪郭線データによる造形物形成時のレーザー照射の軌跡の説明図である。
【図６４】従来の段差の目立たないスライス方法の説明図である。
【図６５】従来の造形物の表面処理による補正方法の説明図である。
【図６６】従来の断面輪郭線データの閉ループの回転方向の是正方法における問題点の説明図である。
【図６７】同上の断面輪郭線データの閉ループの回転方向の是正方法における問題点の説明図である。
【図６８】本発明に用いる光造形形成装置の構成図である。
【図６９】同上の動作説明用フローチャートである。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for creating data for three-dimensional stereolithography used in a three-dimensional shape forming apparatus that cures a photocurable resin in a forming tank layer by layer by laser irradiation.
[0002]
[Prior art]
Conventionally, a three-dimensional forming apparatus in which a photocurable resin in a forming tank is hardened one by one by laser irradiation has the following various problems.
For example, shape data (FIG. 56 (b)) created based on the data shown in FIG. 56 (a) is given to a three-dimensional shape forming apparatus, and a three-dimensional shaped object 1 is formed as shown in FIG. 56 (c). When the ultraviolet light penetrates through the already-cured layer and cures the uncured resin thereunder, the downward portion of the shape of the molded article 1 is excessively cured as shown in FIG. However, there has been a problem that the shape accuracy is greatly reduced. In the figure, ΔH indicates the size of the excessively hardened portion 2, for example, 0.4 to 0.6 mm.
[0003]
Conventionally, a molded object requiring dimensional accuracy has been corrected by manual work after molding. However, not only the number of working steps is increased, but also there are some places where manual correction is not possible depending on the shape.
As a method for solving this point, as disclosed in Japanese Patent Application Laid-Open No. 6-64048, a method has been proposed in which the following correction is performed by an operator's operation at the stage of creating three-dimensional CAD data.
[0004]
That is, in this method, the original shape as shown in FIG. 57A is copied and arranged in the Z-axis direction by the correction amount (ΔH) as shown in FIG. In this method, a Boolean operation is performed on the shape and the offset shape, and only the overlapping portion (the black portion in FIG. 3C) is adopted as the corrected shape.
However, this method also has a problem that even if a designated operation is performed by using a three-dimensional CAD, a calculation error often occurs at the level of a currently commercially available three-dimensional CAD, and a substantially complicated shape cannot be processed. Furthermore, there is a problem that it takes more processing time for the operator and hinders automation of a series of operations.Moreover, since the Boolean operation is used, it can be applied only to solid model data. There was also a problem that we could not cope.
[0005]
Further, there is a problem that processing cannot be performed unless a three-dimensional CAD system for editing the data is possessed.
Further, three-dimensional shape data and cross-sectional contour (contour contour) data exist as input data for formation. However, in the correction for three-dimensional shape data, when the input data is a cross-sectional contour, Cannot be applied, and therefore, it is necessary to realize correction for the cross-sectional contour data.
[0006]
As a correction method for the cross-sectional contour data, a correction value is input as disclosed in Japanese Patent Application Laid-Open No. 8-34064, the number of layers corresponding to the correction value is obtained based on the lamination thickness, and modeling is performed. Sometimes, in each layer other than the layer serving as the bottom surface, in the direction opposite to the direction in which light rays are incident at the time of modeling, the cross-sectional portion where the layer overlaps more than the correction value, or the layer serving as the bottom surface overlaps A method has been proposed in which only one of the cross-sectional portions is corrected to be a cross-section.
[0007]
FIG. 58 shows the flow of the conventional correction method, in which (a) shows the data before correction, (b) shows the data after correction, and (c) shows the actual data based on the data after correction. FIG. 1 shows a modeled object 1 on which a modeling process is performed. According to this correction method, as shown in FIG. 7C, the molded article 1 having a desired shape can be obtained by the excess hardened portion 2 '.
[0008]
However, even if the laser curing conditions are constant, the surplus curing phenomenon of the photocurable resin changes depending on the thickness of the layers to be laminated and the lamination thickness (the number of laminations). Correction can not be realized, the correction may cause a problem of the shape on the contrary, and in order to obtain the correction data of one layer, it is necessary to perform arithmetic processing for the number of layers corresponding to the correction value, There is a problem that the calculation time becomes long.
[0009]
FIG. 59 shows a problem that cannot be solved when a plurality of the above-described lamination pitches (one layer thickness) are used. In order to make the steps on the surface of the modeled product less noticeable, FIG. In this case, modeling data in which the layers are stacked at a fine pitch is used in some cases, but the modeling result is as shown in FIG. 3B, and the problem that the surplus hardened portion 2 becomes uneven because the dimensional deviation differs for each layer. Is shown.
[0010]
Further, data shown in FIG. 60C in which the correction for the excess curing is performed is generated from the data before correction shown in FIG. 60B obtained based on the original data as shown in FIG. Even if modeling is performed based on the data after the correction, a missing portion 3 of the shape occurs as shown in FIG. 3D, and the desired dimensional accuracy cannot be obtained. There was a problem.
[0011]
Furthermore, conventionally, when performing processing of a slice plane on three-dimensional shape data, there is a possibility that cross-sectional contour data that may cause a problem at the time of formation may be generated due to a relationship between a slice coordinate value and a shape.
For example, when forming the three-dimensionally shaped object 1 shown in FIG. 61B having a cross-sectional shape as shown in FIG. 61A, as shown in FIG. When the cross-sectional contour data on the simple slice plane is generated, the cross-sectional contour data as shown in FIG. 62B is generated on the slice plane of (2) due to the relationship between the coordinate value to be sliced and the shape. As shown in FIG. 63 (b), the trajectory at the time of forming an object becomes a double wheel, which causes a problem. FIGS. 62 (a) and (c) are cross-sectional contour data corresponding to the slice planes (1) and (3), and FIGS. 63 (a) and (c) are corresponding to the slice planes (1) and (3). The trajectory corresponding to the cross-sectional contour data to be processed is shown.
[0012]
Furthermore, when forming a three-dimensional structure with a laminated structure, if the laminating pitch is equal pitch (for example, 0.2 mm) as shown in FIG. However, when the lamination pitch is reduced to cope with this problem, there is a problem that the molding time increases. In order to solve this problem, a method of processing unequal pitch slices in which the lamination pitch is partially reduced as shown in FIG. 64B has been proposed. It was necessary to specify the range, which required detailed confirmation of the shape. As a result, the number of oversights increases, and in fact, the modeling is performed once by a test, and the decision is made based on the results. In the case of FIG. 64 (b), the center portion has a pitch of 0.1 mm, and the upper and lower sides have a pitch of 0.2 mm.
[0013]
Further, when there is a defect in the three-dimensional shape data, a defect naturally occurs in the sectional contour data obtained by slicing the data. Generally, when the original data of the three-dimensional shape is a solid model, there are few problems, but when the data is a surface model, there is a problem that the problem is likely to occur. Since the function of a free-form surface is weak in the current solid model, a surface model is often used for a shape having a free-form surface. Also, if the surface model is corrected so that it does not have a defect, it will be reworked, so that the lead time and the modeling man-hour will increase.
[0014]
Further, in a case where the cross-sectional contour data is directly received without CAD data, the correction of the cross-sectional contour data is indispensable because data cannot be corrected by CAD.
Further, when forming a three-dimensional object by laminating thin layers (thin plates), the upward portion of the shape is inside the desired shape and the downward portion of the shape is outside due to data processing (depending on how the data is used). And the opposite is true). FIG. 65 (a) shows the original data in this case, and FIG. 65 (b) shows the modeling result, with the upward part of the original contour being inside the desired shape and the downward part of the shape being inside the desired shape. You can see that it is outside. When a surface treatment is applied to this shaped object to finish it, the upward portion in the case of the polishing process (FIG. 3C) and the downward portion in the case of the coating process (FIG. 3D) are There is a problem that it is difficult to finish the molded object accurately because the error X is further enlarged from the desired shape.
[0015]
Further, when slicing the three-dimensional shape, the direction given to the cross-sectional contour of the closed loop depends on the phase data (information on the front and back of the surface) added to the three-dimensional shape data. When the original data is a solid model, the phase data is correctly added. However, when the surface data is a surface model, the information is not reliable because the original data does not have the phase data.
[0016]
However, the direction of this loop is important when correcting the cross-sectional contour data in order to form a model having accurate dimensions. Although it is necessary to make corrections to improve the forming dimensional accuracy, considering the case where laser spot diameters are generally corrected, it is necessary to offset inward of the shape by the radius of the laser spot diameter. There is. If the left side is determined to be inside the shape with respect to the loop direction, all the processing is offset to the left side in the rotation direction of the loop. Therefore, when the loop direction is correct as shown in FIG. 66A, the laser trajectory data has an offset O corresponding to the radius of the laser spot as shown in FIG. However, when the rotational direction is not good as shown in FIG. 67 (a), the offset is applied in the opposite direction as shown in FIG. 67 (b), and a molded article with accurate dimensions can be obtained. There was no problem.
[0017]
[Problems to be solved by the invention]
The present invention has been made in view of the above-mentioned problems, and has as its object the object of forming data for forming a three-dimensional shaped object using a photocurable resin and post-processing of the shaped object. It is an object of the present invention to provide a method for creating data for three-dimensional stereolithography, which can reduce processing and can form a molded object with high accuracy.
[0018]
[Means for Solving the Problems]
In order to achieve the above object, according to the first aspect of the present invention, a photocurable layer formed by irradiating a photocurable resin with light, or a three-dimensional shape apparatus that forms a desired three-dimensional shape by stacking a plurality of thin plates is provided. In the method for creating data for three-dimensional stereolithography to be used, three-dimensional shape data is formed so as to form a plurality of cross-sectional contour data by slicing the steps so that a step is not conspicuous, and the cross-sectional contour data due to incomplete input shape data. And the step of correcting the cross-sectional contour data to improve the forming dimension accuracy. Thus, the step of slicing and creating a plurality of cross-sectional contour data includes a coordinate value correction process for avoiding slicing the three-dimensional shape data with a defective coordinate value, and a process for making the formed steps inconspicuous. , From the process of slicing the part where the formation step is conspicuous more finely than the normal pitch It is characterized by comprising.
[0020]
Claim 2 In the invention of the claim, 1 In the invention, the slice coordinate value correcting step is performed by calculating the minimum value unit of the moving amount in the stacking direction of the three-dimensional shape forming apparatus for which data is to be created, for each numerical value of the slice coordinate value sequence calculated or designated by the condition. It is characterized in that it is performed by uniformly adding or subtracting any small numerical value less than.
Claim 3 In the invention of the claim, 1 In the invention, the slicing step includes slicing the shape at a constant pitch and determining whether or not a layer for interpolating between two layers formed by the slice is required in a region sandwiched between the two layers. The number of interpolation layers is determined by judging from the normal vector information of the facet of the STL data to be performed, and a slice position is determined based on the number to perform additional slicing between two layers.
[0021]
Claim 4 In the invention of the claim, 3 According to the invention, data having a downward normal vector out of normal vectors of facets of STL data existing in an area between the two layers is excluded from the determination.
Claim 5 In the invention of In a three-dimensional stereolithography data creating method used in a three-dimensional shape device for forming a desired three-dimensional shape by stacking a plurality of thin plates or a photocurable layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects of the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data to The step of correcting the defect of the cross-sectional contour data includes the steps of forming an open loop into a closed loop, dividing and deleting a self-intersection of the closed loop, correcting overlapping loops, and deleting unnecessary closed loops. And correcting the direction of rotation of the closed loop so that a predetermined side of the direction of rotation of the closed loop is inside the shape for data for which it is unclear whether the direction of rotation of the closed loop is correct. .
[0022]
Claim 6 In the invention of the claim, 5 In the invention, in the step of closing the open loop, the open loop attempts to close the loop by itself, and if not closed loop, tries to close the loop by connection with another open loop, If connection is not possible, it is deleted.
Claim 7 In the invention of the claim, 6 In the invention of the above, the method of self-closed loop of the open loop, if there is a self-intersection in the open loop, or if the end point is on the open loop, closed loop at the intersection, delete the rest, and then between the end points If the distance is less than or equal to the specified value, a straight line is added between the end points to form a closed loop, and if the distance between the closed loop and the end point is less than or equal to the specified value, a perpendicular to the closed loop is added from the end point to form a closed loop, It is characterized in that the rest is deleted.
[0023]
Claim 8 In the invention of the claim, 6 or 7 In the invention of the above, the method of forming a closed loop by connecting a plurality of open loops is a method for a case where two open loops have an intersection or a contact point on the condition that the direction of the loop is already set correctly. Rejoin the two open loops at the intersection so that the direction of the loops matches, delete the shorter one, and claim the remaining open loops 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
If the distance between the end point and the start point of another open loop, or the distance between the start point and the end point of another open loop is equal to or less than the specified value, if there are multiple cases, the loop between the shortest distance is the target and between the end points. Connect by adding a straight line and claim the open loop created by this connection 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
For the case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, add a perpendicular from the end point to the loop, and set the loop direction from the intersection of the perpendicular and the loop. Claims about the open loop created by the connection and the deletion by connecting the connection and removing the remaining part by the connection 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
The process of deleting the data of the open loop other than the open loop to be processed in each of the above processes, until the original open loop and the data of the open loop secondaryly generated in the above process disappear. It is characterized in that the processing of the above steps is repeated.
[0024]
Claim 9 In the invention of the claim, 6 or 7 In the present invention, the method of forming a closed loop by connecting a plurality of open loops is performed in a case where a certain two open loops have an intersection or a contact point under a condition that the direction of the loop is unclear. Connecting the long sides divided at an intersection or a contact point, adjusting the direction of the loop to the longer one of the two and removing the rest, and claiming the obtained open loop. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
If the distance between the end point and the loop between other open loops is less than or equal to the specified value, if there is more than one, connect and loop by adding a straight line between the end points for the shortest distance loop To the longer of the two, and claim the resulting open loop. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
For a case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, a perpendicular line from the end point to the loop was added, and divided at the intersection of the perpendicular line and the loop Claim the open loop obtained by connecting with the longer one and adjusting the direction of the loop to the longer of the two and removing the rest. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, it is determined that the processing for both corresponding open loops is completed, and the process proceeds to the next open loop processing, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
The process of deleting the open loop data other than the open loop to be processed in each of the above processes, until the original existing open loop and the data of the open loop secondary generated in the above process disappear, It is characterized in that the processing of the above steps is repeated.
[0025]
Claim 10 In the invention of the claim, 5 In the invention of the present invention, the step of dividing and deleting the self-intersection includes extracting a closed loop at each intersection of the self-intersection and performing closed loop by connecting the remaining loops to finally obtain a plurality of closed loops. It is characterized in that only the closed loop having the largest area is deleted.
Claim 11 In the invention of the claim, 5 In the invention of the above, the process of correcting the overlap of the closed loops is performed by taking the logical sum of the two closed loops when the rotation directions of the two closed loops are the same when the rotation directions of the loops are already set correctly. A closed loop is integrated, and when the rotation direction is different, both closed loops are integrated by subtracting a small-area closed loop from a large-area closed loop.
[0026]
Claim 12 In the invention of the claim, 5 In the invention of the above, the step of deleting the unnecessary closed loop is performed in the case where the rotation direction of the loop has already been correctly set, the inclusion relations of all the closed loops in one cross section are obtained in advance, and the obtained inclusion relation information is used. The uppermost closed loop is set as a parent loop, and the difference between the rotation directions of the parent loop and the closed loop immediately below the parent loop is checked. If the parent loop has the same rotation direction as the parent loop, the closed loop is deleted and deleted. When the lower closed loop is included in the closed loop, the included closed loop is raised to the next higher level, and the closed loop is checked for a difference in rotation direction from the parent loop, and has the same rotation direction as the parent loop. Is characterized in that the operation of deleting the closed loop is performed in order from the parent loop to the lowest closed loop.
[0027]
Claim Thirteen In the invention of the claim, 5 In the invention, the step of correcting the rotation direction of the closed loop includes obtaining the inclusion relation of all the closed loops in one cross section, and using the obtained inclusion relation information as a top-level closed loop as a parent loop with respect to the parent loop. The rotation direction is set such that the inner side becomes the inside of the shape, and the loop rotation direction opposite to the parent loop is set in order to the lowest closed loop.
[0028]
In the invention of claim 14, In a three-dimensional stereolithography data creating method used in a three-dimensional shape device for forming a desired three-dimensional shape by stacking a plurality of thin plates or a photocurable layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects of the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data to The step of correcting the cross-sectional contour data to improve the forming dimensional accuracy includes a loop correcting process according to a surface treatment method after forming and a data correcting process for excess curing of the photocurable resin. And
In the invention of claim 15, the claim 14 In the invention of the present invention, the loop correction process according to the surface treatment method after formation corresponds to a case where surface treatment by polishing is performed after formation, so that the cross-sectional contour data of a certain layer and the lower layer The logical sum with the cross-sectional contour data is used as the correction data of the layer, and the operation of generating the correction data is performed over all the layers.
[0029]
Claim 16 In the invention of the claim, 14 In the invention of the present invention, the loop correction process according to the surface treatment method after the formation corresponds to the case where the surface treatment by coating is performed after the formation, and the cross-sectional contour data of a certain layer and the layer one layer below the layer The logical product with the cross-sectional contour data is used as the correction data of the layer, and the operation of generating the correction data is performed over all the layers.
[0030]
Claim 17 In the invention of the claim, 14 In the invention, in the data correction process for the excess curing of the photocurable resin, the dimensional deviation of the excess curing that varies depending on the condition is preliminarily conditioned and registered in a database, and the dimensional deviation in the correction target layer at the time of the correction. From the database according to the conditions of the thickness and the number of layers of the layer to be corrected and the layer present thereabove, and processing the values searched for, and obtaining the layer below by the amount corresponding to this dimensional deviation. The logical product of the cross-sectional contour data of the layer and the correction target layer is used as the correction data of the correction target layer, and the operation of obtaining the correction data is performed for all layers except the uppermost layer. .
[0031]
Claim 18 In the invention of the claim, 17 In the invention of the above, it is possible to specify a plurality of ranges of layers to be corrected and a range of layers not to be corrected, and to perform correction as independent shape data for each group of one layer range in the layer range to be corrected. Features.
Claim 19 In the invention of the claim, 14 In the invention, the data correction process for the excess curing of the photocurable resin is performed by preconditioning and registering a dimensional deviation of the excess curing that varies depending on conditions in a database, and when the correction is performed, the correction target layer is set to the corresponding layer. Is divided into a plurality of parts according to the condition of the number of layers according to the shape of the plurality of layers existing in the upper layer, the part having no upper layer is used as correction data, and the other parts are dimensional deviations in the respective parts. Is obtained by processing the value searched or retrieved from the database according to the conditions of the thickness and the total number of layers existing above the part and the part concerned, and a layer existing below by an amount corresponding to this dimensional deviation is obtained, The logical product of the cross-sectional contour data of the layer and the portion is used as the correction data of the portion, and the correction data is obtained for all the divided portions. The correction data of the correction target layer by integrating the correction data, and performs over an operation for obtaining the correction data in all layers.
[0032]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described.
FIG. 68 shows a configuration of a three-dimensional shape forming apparatus used in the present invention. The three-dimensional shape forming apparatus includes a light source 10 and a light scanning unit 11 that scans light emitted from the light source 10 on an XY plane. A rectangular frame-shaped resin liquid tank 12 for storing the photo-curable resin liquid 15, a horizontal trapezoidal molding table 13 submerged in the resin liquid tank 12, and a doctor blade that horizontally moves on the resin liquid tank 12. (Sweeping member) 14.
[0033]
The light source 10 is, for example, a gas laser oscillation source that emits ultraviolet laser light, and is mounted on the support stage 20. The light scanning unit 11 is attached to the support stage 20 together with the light source 10. The optical scanning unit 11 includes three fixed mirrors 21, 22, and 23 that reflect laser light from the light source 10, an X scanning mirror 24 that scans light reflected by the fixed mirror 23 in the X direction, and an X scanning mirror 24. And a Y-scanning mirror 25 that scans the laser beam scanned in the X direction in the Y direction and irradiates it toward the resin liquid surface on the molding table 13. The X-scanning mirror 24 and the Y-scanning mirror 25 have respective driving units 26 and 27, and scan the laser light in the respective directions by changing the inclination angles of the driving units 26 and 27. That is, in the optical scanning unit 11, the laser beam 17 projected from the light source 10 is deflected in two directions of X and Y by the two scanning mirrors 24 and 25 and is irradiated on the molding table 13.
[0034]
A rectangular parallelepiped concave portion 12a is formed at one corner of the resin liquid tank 12, and a sensor 30 for extending the irradiation position of the laser light emitted from the light source 10 is provided in the concave portion 12a. Note that a plurality of sensors 30 may be provided instead of one.
The sensor 30 is composed of a two-dimensional position detecting element (PSD), and the amount of deviation in the X and Y directions of the irradiation position of the laser beam irradiated on the sensor 30 from a predetermined position (usually the center of the light receiving surface) is two-dimensionally determined. Can be paid out. The sensor 30 is connected to a computer 31.
[0035]
The computer 31 is also connected to the X and Y driving units 26 and 27, the light source 10, the molding table elevating mechanism, and the doctor blade moving mechanism. The computer 31 controls the movement of the molding table elevating mechanism and the doctor blade moving mechanism, and controls the light source 10 and the X and Y driving units 26 and 27 based on the scanning data. The computer 31 irradiates the sensor 30 with laser light at a predetermined timing, and when the irradiation position is detected by the sensor 30, the computer 31 controls the driving units 26 and 27 based on the amount of deviation from the predetermined position, and , 25 are tilted to correct the variation of the irradiation position.
[0036]
The molding table 13 can be moved up and down in the resin liquid 15 by an elevating mechanism (not shown) installed outside the resin liquid tank 12, and supports the formed photocurable layer 16. The doctor blade 14 is horizontally movable by a horizontal movement mechanism (not shown) installed outside the resin liquid tank 12. The doctor blade 14 sweeps the resin liquid thin film formed on the molding table 13 or the previously formed photocurable layer 16 by horizontal movement to make the thickness of the resin liquid thin layer constant and smooth the surface. It is a member for.
[0037]
An example of the operation of this device will be described with reference to the flowchart shown in FIG. To computer 31 Power supply Is supplied, a predetermined command value is transmitted to the X and Y driving units 26 and 27, and the X and Y scanning mirrors 24 and 25 are tilted and the laser is directed toward the light receiving surface of the sensor 30 in step Sl of FIG. Irradiate light. In step S2, the laser beam and the irradiation position (X0, Y0) of the sensor 30 are read. This irradiation position (X0, Y0) becomes the origin of the subsequent irradiation positions. In step S3, the scanning data is transmitted to the light source 10 and the XY driving units 26 and 27. In step S4, the laser light is scanned in the XY directions on the resin liquid surface. As a result, the exposed portion of the resin liquid 15 on the molding table 13 in the resin liquid tank 12 is hardened to a predetermined shape by one layer, and the photocurable layer 16 is formed. In step S5, the molding table 13 is lowered by a predetermined amount, and then the doctor blade 14 is operated. As a result, a smooth resin liquid thin film having a constant thickness is formed on the formed photocurable layer 16. In step S6, it is determined whether or not the shaping of all the photocurable layers 16 has been completed. If the shaping of all the photocurable layers 16 has been completed, the process ends. If the shaping has not been completed, the process proceeds to step S7. In step S7, a predetermined command value is transmitted to the XY driving units 26 and 27 in the same manner as in step Sl, and the X and Y scanning mirrors 24 and 25 are tilted to irradiate the laser light toward the light receiving surface of the sensor 30. . In step S8, the irradiation position (Xl · Yl) of the laser beam from the sensor 30 is read. In step S9, a correction amount (△ X, △ Y) is obtained from the difference between the current irradiation position (X1 · Y1) and the origin (X0, Y0). In step S10, the scanning data for forming the next photocurable layer 16 is corrected. Specifically, X (scan data after correction) = X (scan data before correction) − △ X, Y (scan data after correction) = Y <scan data before correction) ΔY is used to correct data. . When the correction in step S10 is completed, the process returns to step S3, where the next scan data of the photocurable layer 16 is transmitted to the light source 10 and the XY driving units 26 and 27, and the subsequent operation is repeated until the modeling is completed.
[0038]
The above operation is a basic operation, and the present invention provides a computer 31 of this three-dimensional shape forming apparatus with data such as a three-dimensional CAD which gives data for forming a three-dimensional shaped object. It relates to a data processing method in the generation device 32.
(Embodiment)
The present embodiment relates to a method of creating data in which formation steps are not conspicuous. First, the data generation device 32 shown in FIG. 68 receives three-dimensional shape data including a defect as shown in FIG. Regardless of whether or not there is a defect, correct the slice coordinate value to avoid slicing at the position where the defect occurs, and then slice the three-dimensional shape data so that the formation step is not conspicuous and create multiple cross-sectional contour lines create. As a method of slicing the three-dimensional shape data so that the formation step is not conspicuous, a method as shown in a flowchart in FIG. 2 is employed.
[0039]
The correction of the slice coordinate value is performed based on the flowchart shown in FIG.
That is, it is assumed that the three-dimensional object to be formed is sliced in the Z-axis direction (height direction). When the slice start Z coordinate STARTZ is designated, a minute numerical value α is added to the slice start Z coordinate STARTZ. And the actual slice start Z coordinate STARTZ is a Z coordinate Z that slices the actual slice start Z coordinate STARTZ. _tmp And
[0040]
After this, the last sliced Z coordinate Z _tmp The next time the slice pitch P is added, the Z coordinate Z to be sliced next time _tmp Is determined, and the slice end coordinate ENDZ is changed to the Z coordinate Z. _tmp The slice is sequentially performed until the value exceeds.
For example, when forming the modeled object 1 having a shape as shown in FIG. 4, the height position up to the boundary surface between the lower large-diameter portion 1a and the upper small-diameter portion 1b is set to 30 mm, and the slice pitch P is set. If the designated slice start Z coordinate STARTZ is 0.0, the slice end Z coordinate ENDZ is 50.0, and the minute numerical value α is 0.000333, the designated slice Z coordinate value and the actual slice There is a shift of 0.000333 between the Z-coordinate value and the actual slice Z-coordinate value corresponding to the Z-coordinate value 30.0 at the position where the defect occurs (height position of 30 mm) is 30.000333, which is a defect. No longer occurs.
[0041]
In FIG. 2, after slicing after correcting the slice coordinate value, a portion where the formation step is conspicuous is additionally sliced finer than the normal pitch in order to make the formation step inconspicuous. The method in this process is based on the flowchart shown in FIG.
That is, slices are to be sliced in the Z-axis direction in the same manner as described above, and the normal slice pitch is P, and as shown in FIG. _max STARTZ is the slice start Z coordinate, ENDZ is the slice end Z coordinate, and Z is the Z coordinate to be sliced. _tmp First, processing in this process is performed by setting the slice start Z coordinate STARTZ as Z _tmp And 90 ° is the angle θ between the normal vector of the facet and the z-axis. _min Set to U, Z coordinate Z to slice _tmp While sequentially adding the slice pitch P to the Z coordinate Z _tmp The following processing is performed until the data exceeds the slice end Z coordinate ENDZ.
[0042]
First, the Z coordinate Z _tmp Is read, the facet state, the angle θ between the normal vector of the facet and the z-axis (taken from 0 to 90 °), and the Z coordinate Z to be sliced are read. _tmp To create slice data.
Next, the minimum angle θ in the state II, IV, VI shown in FIG. _min Set D. And the above θ _min U and θ _min The smaller of D is θ _min And Reference numeral 4 denotes a plane to be sliced.
[0043]
Next, from all θ, the minimum angle θ in the state II, III, V _min Set U.
The above θ _min Step a at an angle of _max In this case, the height PP is formed. PP = a _max ・ Tan (θ _min ), And if this PP is PP ≧ P, Z _tmp , And returns to the above processing. When PP is not PP ≧ P, the number of divisions between the lower layer and the lower layer is determined based on the values of PP and P, and the Z value (Z _tmp 1, 2, 3 ...) is calculated. This processing is the additional slice position determination processing
And this Z _tmp After creating 1, 2, 3 ... slice data, Z _tmp Is added to the slice pitch P, and the process returns to the above.
[0044]
Z _tmp Exceeds ENDZ, all slice data are arranged in ascending order of Z coordinate value, and this process is terminated.
Here, the additional slice position determination processing will be described. _max Is 0.5 mm, θ _min Is 5 ° and the slice pitch P is 0.1 mm, the height PP becomes 0.044, and PP <P holds.
[0045]
Then, the number of divisions is determined by rounding up P / PP = 2.2286 and setting the number of divisions = 3. Assuming that the Z value of the lower layer is 100.000333, the middle two Z coordinate values obtained by dividing the interlayer (0.1 mm) into three equal parts are:
Z _tmp 1 = 100.033666
Z _tmp 2 = 100.669999
, And additional slicing is performed at that position.
[0046]
Note that if θ is very close to 0 °, the number of divisions in calculation increases, but the upper limit is set to, for example, up to five divisions in consideration of the mechanical limit of the forming apparatus. Better. In this regard, an angle θ between the normal vector of the facet and the z-axis which is very close to 0 ° (for example, less than 3 °) may be regarded as 0 ° (state I) and excluded from the determination. .
[0047]
When there is input data as described above, the slice coordinate value is corrected to avoid slicing at a position where a defect occurs as shown in FIG. 2, and the formed formation step is made inconspicuous after this correction. After creating data for slicing a conspicuous portion finer than a normal pitch, cross-sectional contour data is generated.
By the way, in a three-dimensional shape forming apparatus using a photocurable resin, the shape of a curing unit by a laser beam is a flat surface on the upper side, but is a spherical shape as shown in FIG. Step 1 is originally not noticeable on the downward surface. Therefore, if the downward surface is not considered, the number of layers can be reduced, and the data processing time and the formation time can be reduced. Further, if the lamination thickness is unnecessarily reduced, the dimensional deviation due to the excessive curing of the photocurable resin increases, and the dimensional accuracy decreases. This is also effective in suppressing this. As the method, the Z coordinate Z to be sliced in FIG. _tmp At the stage of reading all the facet data that crosses the data, the downward data of the normal vector may be skipped.
[0048]
In FIG. 1, three-dimensional shape data is sliced so that the formed steps are not conspicuous, and cross-sectional contour data obtained by creating a plurality of cross-sectional contour lines has a defect in the input three-dimensional shape data. In such a case, a defect also occurs in the cross-sectional contour data.
Therefore, there are the following three conditions for the cross-sectional contour data having no defect.
[0049]
{Circle around (1)} A point where all contour lines are closed (closed loop) instead of an open loop (a loop whose start point and end point do not match) as shown in FIG.
{Circle around (2)} There is no self-intersection (there is a figure-eight portion) as shown in FIG. 9 (b).
{Circle around (3)} As shown in FIG. 9 (c), there is no overlap of loops, and the outlines do not intersect.
Note that the loop has a direction from the start point to the end point, and this direction determines whether the inside of the closed loop is clogged or a hole. In this embodiment, from the start point to the end point of the loop, the right side is determined to be outside the shape and the left side is determined to be inside the shape. According to this rule, when the closed loop is counterclockwise (hereinafter referred to as CCW) Is a solid (solid) state on the inside, and can be said to be a hole (hollow) state when clockwise (hereinafter referred to as CW).
[0050]
FIG. 10 shows a flowchart of a process for correcting a defect included in data. In this flowchart, a process of making an open loop into a closed loop, a process of dividing and deleting a self-intersection of the closed loop, and correcting an overlap of the closed loop are performed. Through the process, the above conditions (1) to (3) are satisfied, and a process of deleting unnecessary closed loops and a process of correcting the direction of rotation of the closed loop are performed to generate cross-sectional contour data in which a defect is corrected. You do it.
[0051]
Here, the process of making the open loop closed is performed according to the flowchart shown in FIG.
That is, in this process, all the data of the open loop is read, and the closed loop forming process by itself, the closed loop forming process by connection with another open loop, and the process of deleting the open loop that cannot be connected are performed.
[0052]
The closed loop processing by itself is performed according to the flowchart shown in FIG. That is, the open loop data is sequentially selected one by one, and the selected closed loop has an intersection between the start point S and the end point E as shown in FIG. Alternatively, when an end point (start point S or end point E) exists on the loop as shown in FIG. 14A, a closed loop is formed at the intersection as shown in FIG. 13B or FIG. As shown in FIG. 13 (c) or FIG. 14 (c), a process for matching the start point S and the end point E is performed.
[0053]
As shown in FIG. 15A, when an open loop exists in which the distance G between the end points (between the start point S and the end point E) is equal to or smaller than the specified value ΔG, as shown in FIG. A straight line is added between the end points to form a closed loop as shown in FIG.
Further, when the distance G from the end point (start point S or end point E) is equal to or smaller than the specified value ΔG as shown in FIG. 16A, a perpendicular line from the end point to the loop is shown in FIG. 16B. As shown in FIG. 16C, a closed loop is formed. When the closed-loop processing is completed for the open-loop data that can be closed by itself in this way, the process exits from the closed-loop processing by itself.
[0054]
The data generated by the closed loop is registered in the data processing device 32 and the original closed loop data that has been closed is deleted.
In the case where the closed loop cannot be performed by itself and the closed loop is formed by connection with another open loop, the closed loop processing is performed according to the flowcharts shown in FIGS. The case of FIG. 17 is used when the direction of the loop is already set correctly.
[0055]
In this case, when the open loop data is sequentially selected one by one, and as shown in FIG. 18A, the open loop data having the intersection or the contact with the open loop having the intersection or the contact with another open loop is selected. 18 (b), the two loops are connected at the intersection in the same direction, and the shorter one is deleted as shown in FIG. 18 (c). When the distance between the end point E and the start point S of the other open loop and the distance between the start point S and the end point E of the other open loop are selected as shown in FIG. A process of adding a straight line between the end points as shown in FIG. 19B and connecting both loops as shown in FIG. 19C is performed.
[0056]
When the distance between the end point and the other open loop and the distance between the loop and the end point of the other open loop are selected as shown in FIG. 20B, a process is performed to add a perpendicular line to the direction which is divided at the intersection with the perpendicular line as shown in FIG. 20C.
After connecting a plurality of open loops and performing a closed loop processing as described above, further perform a closed loop processing by itself in the same manner as described above, and when the closed loop processing is finally completed, the closed loop processing is performed. The generated closed-loop data is registered, and the original two closed-loop open-loop data are deleted. If the closed loop cannot be formed, the connected open loop data is deleted, and the data obtained by the connection is added as the new open loop data.
[0057]
Then, if the open loop data selected in the process is open loop data that does not correspond to any of the above cases, the open loop data is deleted.
In this way, the corresponding open loop data is sequentially selected, and when the selection and processing of all data are completed, the process exits from the process.
In addition, in the closed loop processing for connecting a plurality of open loops, if it is not clear whether the handling of the open loop is correct or not, the closed loop processing is performed according to the flowchart of FIG.
[0058]
In this case as well, the open loop data is sequentially selected one by one, and as shown in FIG. 22 (a), when the open loop data having an intersection or a contact with an open loop having an intersection or a contact with another open loop is selected. Then, as shown in FIG. 22 (b), the longer sides divided at the intersections are connected, the loop direction is adjusted to the longer one, and the rest is deleted as shown in FIG. 22 (c).
[0059]
When the distance between the end point E and the start point S of the other open loop and the distance between the start point S and the end point E of the other open loop are selected as shown in FIG. A process of adding a straight line between the end points as shown in FIG. 23 (b) and adjusting the loop direction to the longer one, and connecting both loops as shown in FIG. 23 (c) is performed.
Further, if the distance between the end point and another open loop and the distance between the loop and the end point of the other open loop are selected as shown in FIG. 24B is added as shown in FIG. 24 (b), connected to the longer direction divided at the intersection with the perpendicular, and the loop direction is adjusted to the longer direction, and as shown in FIG. Is performed.
[0060]
After connecting a plurality of open loops and performing a closed loop processing as described above, further perform a closed loop processing by itself in the same manner as described above, and when the closed loop processing is finally completed, the closed loop processing is performed. The generated closed-loop data is registered, and the original two closed-loop open-loop data are deleted. If the closed loop cannot be formed, the connected open loop data is deleted, and the data obtained by the connection is added as the new open loop data.
[0061]
Then, if the open loop data selected in the process is open loop data that does not correspond to any of the above cases, the open loop data is deleted.
In this way, the corresponding open loop data is sequentially selected, and when the selection and processing of all data are completed, the process exits from the process.
Now, in FIG. 10, after the process of forming an open loop into a closed loop by the above-described method is completed, a process of dividing / eliminating the self-intersection of the closed loop is performed. In this process, the process according to the flowchart shown in FIG. 25 is performed. I do.
[0062]
That is, in this case, the closed loop data is sequentially selected one by one, and if the selected closed loop has a self-intersection, the closed loop is cut out at the intersection and divided into two closed loops.
In these divisions, the presence or absence of self-intersection is determined again using the closed loop data, and if there is, the closed loop is cut out again at the intersection and divided into two closed loops. In this way, the division processing is performed until there is no self-intersection, and if the division processing divides the data into a plurality of closed loops, only the closed loop having the largest area in the closed loops is registered as new data, and the others are deleted. A process for deleting the original closed loop data is performed.
[0063]
FIG. 26A shows a closed loop having two self-intersections. In the case of this closed loop, it is divided into three closed loops by a dividing process (FIG. 26B), and the largest closed loop is shown in FIG. 26 (c).
In this way, the corresponding closed loop data is sequentially selected, and when all the data are selected and processed, the process exits from the process.
[0064]
When the process of dividing / deleting the self-intersection of the closed loop is completed, the process proceeds to the process of correcting the overlap of the closed loop as shown in FIG. 10. In this process, the process shown in the flowchart of FIG. 27 is performed. However, this corresponds to a case where the rotation direction of the loop has already been correctly set.
In other words, in this case, the closed loop data is sequentially selected one by one, and it is determined whether or not the selected closed loop overlaps another closed loop. In some cases, it is further determined whether the rotation direction of the loop is the same or different. When the rotation directions (phases) of the loops are equal as shown in FIG. 28A, the two closed loops are ORed to integrate the loops as shown in FIG. When the rotation directions (phases) of the loops are different as shown in FIG. 29 (a), they are integrated as shown in FIG. 29 (b) by subtracting a closed loop with a small area from a closed loop with a large area. Add loop data and delete the original closed loop data.
[0065]
In this way, the corresponding closed loop data is sequentially selected, and when all the data are selected and processed, the process exits from the process.
When this process is completed, an unnecessary closed-loop elimination process starts as shown in FIG. 10. In this process, the process shown in the flowchart of FIG. 30 is performed. However, in this case, it is assumed that the rotation direction of the loop has already been set correctly.
[0066]
First, the inclusion relations of all the closed loops in one cross section are checked and set in the memory of the data generating device 32.
In this setting method, when there is each of the loops LP1 to LP6 as shown in FIG. 31A, the level of the inclusion relation and the rotation direction of the loop are collectively set in a table as shown in Table 1.
[0067]
[Table 1]

[0068]
Then, data of a loop number, a level, and a rotation direction are sequentially read from the first data, and the rotation direction is stored in a work area WORK (level 0). Then, it is determined whether or not the level of the data is 0. Here, since the level of the loop LP1, which is the first data, is 0, it is determined to be a parent loop, the next second data is read, and the rotation direction is written in the work area WORK (level 1).
[0069]
Here, since the loop LP2 as the second data is at level 1, it is next determined whether or not the rotation direction is equal to the parent loop LP1 at level 0.
Here, since it is different here, the data of the third loop LP4 is read out next, and the rotation direction is written in the work area WORK (level 2). In this case, since the level is 2, the rotation direction is compared with the rotation direction of the level 1 loop. Here, since the rotation direction of the second loop LP2 is different from the rotation direction of the loop LP4, the data of the next fourth loop LP5 is read, and the rotation direction is written in the work area WORK (level 3) corresponding to the level 3. . Here, since the level is level 3, the rotation direction of the loop LP4 of level 2 is compared with the rotation direction. In this case, since the rotation direction is different, the data of the loop LP3, which is the fifth data, is read, and the rotation direction is written in the work area WORK (level 1) corresponding to the level 1. Here, since the level is 1, it is compared with the rotation direction of the loop LP1 of level 0. In this case, since the rotation directions of both the loops LP1 and LP3 are the same, the data of the fifth loop number is deleted.
Then, the level of the parent loop, that is, the original, is set as the data of the currently called loop, and then the sixth data, that is, the data of the loop LP6, is read. If the reading of all data has not been completed by this, The process proceeds to the reading of the next seventh data. In the case of FIG. 31A, the process ends. Therefore, the level of the loop LP6 is compared with the original level. That is, the level of the loop LP is set to 1 by subtracting 1 from 2.
[0070]
In this way, even for defective data, unnecessary loops can be determined and deleted according to the rotation direction of the loop as shown in FIG. 31 (b). Can be reduced.
When the unnecessary closed loop elimination process is completed, the process proceeds to the process of correcting the rotation direction of the closed loop as shown in FIG. In the correction of the rotation direction of the closed loop, a process according to a flowchart shown in FIG. 32 is performed. In this case, the inclusion relations of all the closed loops in one section are checked and the loop number and the level are set in the memory, as described above. Here, the case of FIG. 31A is taken as an example.
[0071]
First, the first data is sequentially read from the set data, and the data is read to see whether the data level is odd or even. The data of the loop number of the odd number is converted into the rotation direction CW. Is converted to the rotation direction CCW.
In this way, even if the rotation direction of the loop is unreliable, the inside and outside are determined based on the inclusion relation of the loop in the cross section, and the rotation direction is corrected accordingly, thereby preventing the formation failure caused by this, Data modification man-hours can be reduced.
[0072]
As described above, the process of the defect correction shown in FIG. 10 is completed, and the data generating device 32 creates the cross-sectional contour data after the defect correction, and as shown in FIG. Enter the stage of correcting the data.
At this stage, as shown in FIG. 33, there are data correction processing according to the method of surface treatment after formation, and data correction processing to correct excessive curing of the photocurable resin. In the case of the data correction processing according to the method, there are a correction method when performing a surface treatment by polishing after formation and a correction method when performing a surface treatment by coating after formation.
[0073]
In the former correction method, the processing according to the flowchart shown in FIG. 34 is performed.
In other words, the total number of laminated structures is N _max Then, the logical sum of the n-th layer and the (n-1) -th layer, which is the preceding layer, is calculated, and the process is performed as the correction data of the n-th layer. However, the data of the 0th layer is cross-sectional contour data sliced at the lowest Z value of the three-dimensional shape. Therefore, when the data before the correction is as shown in FIG. 35A, the data after the correction is the shape data located outside the desired shape in both the upward and downward portions as shown in FIG. 35B. After forming the modeled object based on the corrected shape data, a desired dimension can be ensured only by polishing the outer portion, and the modeled object 1 of accurate dimensions can be obtained by an easy operation as shown in FIG. Will be obtained.
[0074]
The desired dimensions can be ensured only by polishing the formed steps after forming the molded article using the corrected data as described above, and the molded article 1 having accurate dimensions can be obtained by an easy operation.
In the latter correction method, the processing according to the flowchart shown in FIG. 36 is performed.
In other words, the total number of laminated structures is N _max Then, the logical product of the n-th layer and the (n-1) -th layer preceding the n-th layer is calculated to obtain the n-th layer correction data. However, the data of the 0th layer is the cross-sectional contour data sliced at the lowest Z value of the three-dimensional shape.
[0075]
Therefore, when the data before the correction is as shown in FIG. 37A, the data after the correction is the shape data in which the downward portion located outside the desired shape is also located inside as shown in FIG. 37B. After forming the modeled object based on the corrected shape data, the stepped portion is filled with a coating process, whereby a desired size can be secured. ).
[0076]
After performing the data correction corresponding to the polishing or the coating process as described above, the data correction process for correcting the excessive curing of the photocurable resin is performed.
As a data correction process for correcting the excessive curing of the photocurable resin, there is the following method.
First of all, a dimensional deviation table corresponding to a condition is registered in advance as a database by experiment or computer analysis. Table 2 shows an example of this database. In this case, the dimensional deviation from the first layer to the fifth layer is shown, and the same value is set for the fifth layer and above.
[0077]
[Table 2]

[0078]
This process is performed according to the flowchart shown in FIG.
That is, the layer N corresponding to the maximum number of layers _max The correction process is sequentially performed from the lower layer to the lower layer. That is, the dimensional deviation of the n-th layer is calculated from the condition and the numerical value of the database, and the layer number of the layer below the dimensional deviation is obtained. However, when the position indicated by the calculated numerical value indicates the middle of a certain layer, the layer number of the next higher layer is obtained. If the layer number m is not larger than the 0th layer, the correction data of the nth layer is empty data (no data), and if the layer number m is larger than the 0th layer, the logical product operation of the nth layer and the mth layer is performed. Then, the correction data of the n-th layer is obtained.
[0079]
This correction process will be described in detail based on the shape data in FIG. In the shape data in FIG. 39, the maximum number of layers is 100, and the thickness of each of the 100 to 96 layers is 0.2 mm, and the thickness of each of the 95 to 86 layers is 0.2 mm.
(1) Since the 100th layer at the top is the top layer, no correction is performed.
(2) 99th layer Two layers of 0.2 mm, with dimensional deviation of 0.25 mm and 0.25 mm below, are the 98th layer.
(3) 98th layer Three layers of 0.2 mm, with dimensional deviation of 0.3 mm and 0.3 mm below are the 97th layer.
(4) 97th layer Four layers of 0.2 mm, with dimensional deviation of 0.33 mm and 0.33 mm below are the 95th layer.
(5) 96th layer Five layers of 0.2 mm, with dimensional deviation of 0.35 mm and 0.35 mm below are the 93rd layer
(6) 95th layer One layer of 0.1 mm and five layers of 0.2 mm, the dimensional deviation is 0.25 mm + (0.35−0.15) = 0.45 mm, and the lower layer of 0.45 mm is 91 layers. Eye
(7) 94th layer Two layers of 0.1 mm and five layers of 0.2 mm have a dimensional deviation of 0.4 mm + (0.35−0.25) = 0.5 mm, and 89 layers below 0.5 mm. Eye
(8) 93rd layer The dimensional deviation is 0.5mm + (0.35-0.3) = 0.55mm with 3 layers of 0.1mm and 5 layers of 0.2mm, and 88 layers are below 0.55mm. Eye
(9) 92nd layer The dimensional deviation is 0.53mm + (0.35-0.33) = 0.55mm with 4 layers of 0.1mm and 5 layers of 0.2mm, and 87 layers below 0.55mm. Eye
(10) 91st layer Five layers of 0.1 mm, the dimensional deviation is 0.55 mm, and the lower part of 0.55 mm is the 86th layer.
In the same manner, the logical product (overlapping portion) of the two layers, that is, the layer to be corrected and the obtained lower layer is calculated as the correction data of the layer to be corrected.
[0080]
FIG. 40 shows a flow in a case where a molded object is formed by performing data correction. The corrected data shown in FIG. 40B is obtained from the formation data shown in FIG. Thus, it can be seen that, when the molded article 1 is formed, the molded article 1 having a shape close to a desired shape can be obtained even when there is an excess cured portion X.
In other words, even when the thickness of one layer differs depending on the portion, which is impossible in the related art, it is possible to correct the excess curing of the photocurable resin. Further, even if the conventional correction is performed, even if the uppermost portion has a cup-shaped open shape, which is disadvantageous, it can be corrected to an accurate shape. Further, when a certain layer is corrected, the logical product operation needs to be performed only once, so that the processing time is reduced.
[0081]
In the processing shown in FIG. 38, in the case of the shape shown in FIGS. 41 (a) to 41 (d), a shape missing portion is generated at the uppermost end of the downward surface as shown in FIG. May decrease. 41A shows the original shape data, FIG. 41B shows the shape data before correction having a defect, FIG. 41C shows the data after correction, and FIG. 41D shows the formed object 1 based on the data after correction. Indicates the status.
[0082]
To compensate for this, the processing method shown in FIG. 42 allows a plurality of layer ranges not to be corrected and a plurality of layer ranges to be corrected to be specified. That is, a correction target range is designated and input, and it is determined whether or not the layer to be processed is the target range, and if not, the correction is canceled.
This will be described with reference to specific examples shown in FIGS. 43A to 43D. In the data including the defect (FIG. 43B) obtained from the original data in FIG. The data correction is performed by designating the range to be corrected and the shape data as shown in FIG. 43 (c) is created, and the modeled object 1 is shown in FIG. 43 (d) based on the corrected shape data. With such a configuration, the missing portion is eliminated, and even in a portion having a step, a shape defect due to the correction is suppressed, and the molded article 1 having an accurate shape can be obtained.
[0083]
FIGS. 44 (a) to 44 (d) show another example, in which a range to be corrected from data (FIG. 44 (b)) containing a defect obtained from the original data of FIG. 44 (a). Is divided into two parts, and correction is performed in each of them. When the modeled object 1 is formed as shown in FIG. 44 (d) based on the corrected data (FIG. 44 (c)), the missing portion is formed in the same manner as in the above case. Is eliminated, and even in a portion having a step, the problem of the shape due to the correction is suppressed, and the modeled object 1 having an accurate shape can be obtained.
[0084]
As described above, by properly supporting the range of the layer to be corrected and the range of the layer not to be corrected, even if the shape has a step, the problem of the shape due to the correction is suppressed, and the modeling of the accurate shape is performed. Object 1 can be obtained.
By the way, the correction processing of the surplus hardening shown in FIG. 38 or FIG. 39 can perform correction by a relatively easy procedure and calculation, but does not guarantee accurate correction for all shapes.
[0085]
That is, in these processing methods, the correction for a certain layer is performed under the same conditions over the entire area of the layer. In the actual shape, even in one layer, the dimensional deviation of the excess hardening may be different due to the different conditions of the layer overlying the partial portion. Correct correction of the shape as shown in FIG. 45 cannot be performed.
That is, the original data having the shape as shown in FIG. 45A is converted into the uncorrected data having the defect shown in FIG. 45B, and the data is further corrected by the processing method of FIG. When data as shown in c) is created and the building 1 is formed from this data, a missing portion is generated as shown in FIG. On the other hand, data correction is performed by designating a correction range as shown in FIG. 45E by the processing method shown in FIG. 42, and the modeled object 1 is formed with the corrected data as shown in FIG. Has an excessively shaped portion.
[0086]
The method described below is a method of correcting the excess hardening for all shapes in consideration of such points.
In other words, in this method, for each laser scanning condition, a dimensional deviation that differs depending on the thickness and the number of layers is obtained in advance by experiment or computer analysis and is stored in a database in advance.
[0087]
Then, the processing of the flowchart shown in FIG. 46 is executed. In this process, when performing correction of each layer as shown in the figure, the layer to be corrected is divided according to the condition of the number of overlapping layers depending on the shape of a plurality of layers existing above the layer, and the size in each part is The deviation is calculated from the numerical value of the database (same as in Table 2) according to the condition of the relevant portion, and then the lower layer corresponding to the correction value is obtained, and further, the corresponding layer and the corresponding portion of the layer to be corrected are calculated. Shapes that overlap with the shapes are obtained, and integrated with those shapes to obtain correction data for the layer to be corrected.
[0088]
The correction data to be obtained is obtained by performing this over all layers. However, no correction is made for the uppermost part.
This correction processing will be described for a more specific shape.
When correcting the shape data shown in FIG. 47 in which the top layer is the 100th layer and the layer of the stepped portion is 89 layers,
(1) Since the 100th layer at the top is the top layer, no correction is performed.
[0089]
(2) 99th layer The two layers of 0.1 mm have a dimensional deviation of 0.4 mm and the lower part of 0.4 mm is the 95th layer, which is the logical product.
(3) 98th layer The three layers of 0.1 mm have a dimensional deviation of 0.5 mm and the lower part of 0.5 mm is the 93rd layer, which is the logical product.
(4) 97th layer The four layers of 0.1 mm have a dimensional deviation of 0.53 mm and the lower part of the layer is 0.53 mm.
[0090]
(5) 96th layer The five layers of 0.1 mm have a dimensional deviation of 0.55 mm and the lower part of 0.55 mm is the 91st layer, which is the logical product.
(6) Similarly, the 95th to 90th layers are logically ANDed with the 90th to 85th layers.
The conditions are the same for the 100th to 90th layers because the conditions are the same over the layers and the layers are not divided. Similar to the processing method of FIG. 38, the logical product (overlapping part) of the two layers of the correction target layer and the lower layer obtained is obtained. ) Is calculated as correction data of the correction target layer.
[0091]
(7) In the 89th layer, as shown in FIG. 48 (a), the portion (1) is the topmost, but the portion (2) is a 0.1-mm five-layer structure with a dimensional deviation of 0.55 mm. AND with FIG. 48B shows the correction result of the 89th layer.
(8) As shown in FIG. 49A, part (1) of the 88th layer has two layers of 0.1 mm, a dimensional deviation of 0.4 mm, and the logical product of the 84th layer. Further, the portion (2) has 0.1 mm and 5 layers, and has a dimensional deviation of 0.55 mm, which is the logical product of the 83rd layer. FIG. 49B shows a correction result of the 88th layer.
[0092]
Hereinafter, correction is performed in the same manner.
FIG. 50 (a) shows the corrected shape data, and FIG. 50 (b) shows the result of forming the building 1 using this shape data. As a result, the missing and excessive portions of the shape are eliminated. , The exact shape.
Although the shape used in the above description is a simplified shape for the purpose of explanation, any shape can be processed in the same way.
[0093]
As described above, according to the processing method shown in FIG. 46, since the dimensional deviation due to the excessive curing can be corrected for each portion, accurate correction can be realized for any shape.
The shape division of each layer is performed by the following method.
For example, when the shape viewed from above is the shape shown in FIG. 51 (a) and the shape viewed from the side is the shape shown in FIG. 51 (b), as shown in FIG. Of the difference between Lt and the upper layer L1, a portion A1 not in the upper layer is divided.
[0094]
That is, A1 (the part to be processed as the uppermost layer) is the part of the guarantee target layer Lt where the layer L1 does not overlap, and is processed as the uppermost layer. In the drawing, B1 indicates the remaining portion of the correction target layer Lt where the layer L1 overlaps.
Next, as shown in FIG. 52 (b), of the difference between the remaining portion B1 and the two-layer upper layer L2, a portion A2 not in the upper layer is divided. In the drawing, B2 indicates the remaining portion of the correction target layer Lt overlapping the layer L2.
[0095]
In the same manner, the division is performed by repeating the operation of cutting out the non-overlapping part of the upper layer from the difference between the remaining part and the higher layer.
In this case, since the dimensional deviation becomes a constant value for a certain degree or more, it is sufficient to perform the upper three or four layers.
FIG. 53A shows a state in which, of the difference between the remaining portion B2 and the three upper layers L3, the portion A3 of the correction target layer Lt where the layer L3 does not overlap is divided. In the drawing, B3 indicates the remaining portion of the correction target layer Lt overlapping the layer L3.
[0096]
FIG. 53B shows a state in which, of the difference between the remaining portion B3 and the three upper layers L4, the portion A4 of the correction target layer Lt where the layer L4 does not overlap is divided. In the drawing, B4 indicates the remaining portion of the correction target layer Lt overlapping the layer L4.
The result of the above division is as shown in FIG. 54, and these portions may be corrected according to the respective conditions.
[0097]
Next, for each partial shape in this example, A1 is uncorrected, A2 is the logical product of one layer below, A3 is the logical product of two layers below, A4 is the logical product of three layers below, and A5 is the logical product of four layers below. As shown in FIG. 55 (a), the correction result is obtained for each partial shape as shown in FIG. 55 (a). When these partial shape results are integrated, the result becomes as shown in FIG. 55 (b). The portion shown in FIG. 55B (a) shows the layer shape after correction, and the layer shape before correction shows the entire rectangle (b).
[0098]
【The invention's effect】
The invention according to claim 1 is a three-dimensional stereolithography data used in a three-dimensional shape device that forms a desired three-dimensional shape by stacking a plurality of thin plates or a photocurable layer formed by irradiating light to a photocurable resin. In the creating method, a step of forming a plurality of cross-sectional contour data by slicing the three-dimensional shape data so that a step is not conspicuous, and a step of correcting a defect of the cross-sectional contour data due to a defect in the input shape data. And the step of correcting the cross-sectional contour data in order to improve the forming dimension accuracy, thereby reducing the man-hours for creating the data for forming the three-dimensional shape and the post-processing of the formed object, reducing the cost of the three-dimensional shape forming work, There is an effect that the delivery time can be shortened and the precision of the formed object can be improved.
[0099]
Also, The step of creating the cross-sectional contour line data by slicing includes a coordinate value correction process for avoiding slicing the three-dimensional shape data with a defective coordinate value, and a location where the formation step is conspicuous in order to make the formation step inconspicuous. And a step of slicing the finer than the normal pitch, so that it is possible to obtain a shaped article with less noticeable level difference, and thus to reduce the post-processing of the shaped article.
[0100]
Claim 2 The invention of claim 1 In the invention, the slice coordinate value correcting step is performed by calculating the minimum value unit of the moving amount in the stacking direction of the three-dimensional shape forming apparatus for which data is to be created, for each numerical value of the slice coordinate value sequence calculated or designated by the condition. Since it is performed by uniformly adding or subtracting any small numerical value less than, when slicing the shape data and performing data correction, it is possible to prevent slicing at the position where a defect occurs, preventing formation defects due to this, This has the effect of reducing the number of data correction steps.
[0101]
Claim 3 The invention of claim 1 In the invention, the slicing step includes slicing the shape at a constant pitch and determining whether or not a layer for interpolating between two layers formed by the slice is required in a region sandwiched between the two layers. The number of interpolation layers is determined by judging from the normal vector information of the facet of the STL data to be performed, the slice position is determined based on the numerical value, and additional slicing is performed between the two layers. The slice data can be created by automatically recognizing the slice data, so that there is no need to overlook or form a test, and it is possible to form a shaped object with no noticeable step without increasing the number of work steps.
[0102]
Claim 4 The invention of claim 3 In the invention of the above, among the normal vectors of the facet of the STL data existing in the region sandwiched between the two layers, the data having the downward normal vector is excluded from the judgment, so that the downward surface where the step is not originally noticeable is considered. Since the correction is not performed, the number of laminations can be reduced by that much, and the data processing time and the formation time can be shortened.In addition, since the lamination thickness is not reduced, the dimensional deviation caused by the excessive curing of the photocurable resin is reduced. There is an effect that a reduction in dimensional accuracy can be suppressed.
The invention of claim 5 is In a three-dimensional stereolithography data creating method used in a three-dimensional shape device for forming a desired three-dimensional shape by stacking a plurality of thin plates or a photocurable layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects of the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data to The step of correcting the defect of the cross-sectional contour data includes the steps of forming an open loop into a closed loop, dividing and deleting a self-intersection of the closed loop, correcting overlapping loops, and deleting unnecessary closed loops. And the process of correcting the closed-loop rotation direction so that the predetermined side of the closed-loop rotation direction is inside the shape for data for which it is unclear whether the closed-loop rotation direction is correct. Even if it is a section profile data, it is automatically judged and converted to a section profile data without any problem. In addition to being able to reduce the cost of the three-dimensional shape forming work, shorten the delivery time, and improve the precision of the molded object, This has the effect of significantly reducing the number of data correction steps.
[0103]
Claim 6 The invention of claim 5 In the invention, in the step of closing the open loop, the open loop attempts to close the loop by itself, and if not closed, attempts to close the loop by connecting with another open loop, and performs the other open loop. If the connection cannot be made, it is deleted. Therefore, it is possible to close the loop in which the cross-sectional contour data is open loop to eliminate the defect, and to delete the cross-sectional contour data that does not solve the problem and remove its influence. effective.
[0104]
Claim 7 The invention of claim 6 In the invention of the above, the method of the closed loop of the open loop by itself, if there is a self-intersection in the open loop or if the end point is on the open loop, closed loop at the intersection, delete the rest, and then between the end points If the distance is less than or equal to the specified value, a straight line is added between the end points to form a closed loop, and if the distance between the closed loop and the end point is equal to or less than the specified value, a perpendicular to the closed loop is added from the end point to form a closed loop, Since the remainder is deleted, there is an effect that a plurality of open loops can be closed.
[0105]
Claim 8 According to a sixth aspect of the present invention, in the method of the sixth or seventh aspect, the method of forming a closed loop by connecting the plurality of open loops is performed under the condition that the direction of the loop is already set correctly. When there is an intersection or a contact point, the two open loops are connected at the intersection so that the directions of the loops match, the shorter one is deleted, and the remaining open loops are claimed. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
If the distance between the end point and the start point of another open loop, or the distance between the start point and the end point of another open loop is equal to or less than the specified value, if there are multiple cases, the loop between the shortest distance is the target and between the end points. Connect by adding a straight line and claim the open loop created by this connection 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
For the case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, add a perpendicular from the end point to the loop, and set the loop direction from the intersection of the perpendicular and the loop. Claims about the open loop created by the connection and the deletion by connecting the connection and removing the remaining part by the connection 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
The process of deleting the data of the open loop other than the open loop to be processed in each of the above processes, until the original open loop and the data of the open loop secondaryly generated in the above process disappear. Since the process of the above process is repeated, there is an effect that a plurality of open loops can be made into a closed loop under the condition that the direction of the loop is already correctly set.
[0106]
Claim 9 The invention of claim 6 or 7 In the invention of the present invention, a method of forming a closed loop by connecting a plurality of open loops is performed in a case where a certain two open loops have an intersection or a contact point under a condition that the direction of the loop is unclear. Connecting the long sides divided at an intersection or a contact point, adjusting the direction of the loop to the longer one of the two and removing the rest, and claiming the obtained open loop. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
If the distance between the end point and the loop between other open loops is less than or equal to the specified value, if there is more than one, connect and loop by adding a straight line between the end points for the shortest distance loop To the longer of the two, and claim the resulting open loop. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
For a case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, a perpendicular line from the end point to the loop was added, and divided at the intersection of the perpendicular line and the loop Claim the open loop obtained by connecting with the longer one and adjusting the direction of the loop to the longer of the two and removing the rest. 7 The closed loop is tried by the method of the closed loop of the open loop by itself, and if the closed loop is formed by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop, and the trial is performed. If not closed loop in the process of adding the data of the open loop to the end of the processing data,
The process of deleting the open loop data other than the open loop to be processed in each of the above processes, until the original existing open loop and the data of the open loop secondary generated in the above process disappear, Since the process of the above process is repeated, there is an effect that even if the direction of the loop is unclear, there is an effect that it is possible to cope with the closed loop of a plurality of open loops.
[0107]
Claim 10 The invention of claim 5 In the invention of the present invention, the step of dividing and deleting the self-intersection includes extracting a closed loop at each intersection of the self-intersection and performing closed loop by connecting the remaining loops to finally obtain a plurality of closed loops. Since only the closed loop having the largest area is deleted, the self-intersection of the closed loop can be automatically divided and deleted.
[0108]
Claim 11 The invention of claim 5 In the invention of the above, the step of correcting the overlap of the closed loops is performed by taking the logical sum of the two closed loops when the rotation directions of the two closed loops are the same when the rotation directions of the loops are already set correctly. When the closed loops are integrated and the rotation direction is different, both closed loops are integrated by subtracting the small area closed loop from the large area closed loop. The effect is that it can be done automatically.
[0109]
Claim 12 The invention of claim 5 In the invention of the above, the step of deleting the unnecessary closed loop includes, in a case where the rotation direction of the loop has already been set correctly, previously obtaining the inclusion relations of all the closed loops in one cross section, and using the obtained inclusion relation information. The uppermost closed loop is set as a parent loop, and the difference between the rotation directions of the parent loop and the closed loop immediately below the parent loop is checked. If the parent loop has the same rotation direction as the parent loop, the closed loop is deleted and deleted. If the lower closed loop is included in the closed loop, the included closed loop is raised to the next higher level, and also the closed loop is checked for a difference in rotation direction from the parent loop, and has the same rotation direction as the parent loop. Performs the operation of deleting the closed loop in order from the parent loop to the lowest closed loop, so that an unnecessary closed loop can be created under the condition that the rotation direction of the loop is set correctly. There is an effect that can be deleted dynamically.
[0110]
Claim Thirteen The invention of claim 5 In the invention, the step of correcting the rotation direction of the closed loop includes obtaining the inclusion relation of all the closed loops in one cross section, and using the obtained inclusion relation information as a top-level closed loop as a parent loop with respect to the parent loop. The rotation direction is set so that the inside is inside the shape, and the loop rotation direction opposite to the parent loop is set in order until the lowest closed loop, so even if the rotation direction of the loop is not reliable, By judging inside / outside based on the inclusion relation of the loop and correcting the rotation direction in accordance therewith, it is possible to prevent the formation trouble caused by this and to reduce the number of data correction man-hours.
[0111]
The invention of claim 14 is In a three-dimensional stereolithography data creating method used in a three-dimensional shape device for forming a desired three-dimensional shape by stacking a plurality of thin plates or a photocurable layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects of the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data to Since the step of correcting the cross-sectional contour data to improve the forming dimensional accuracy includes a loop correcting process according to the surface treatment method after forming and a data correcting process for excess curing of the photocurable resin, In addition to being able to reduce the cost of the three-dimensional shape forming work, shorten the delivery time, and improve the precision of the molded object, There is an effect that data correction can be performed to improve the formation dimensional accuracy corresponding to the surface treatment method performed after the formation of the modeled object.
[0112]
Claim Fifteen The invention of claim 14 In the invention of the present invention, the loop correction process according to the surface treatment method after formation corresponds to a case where surface treatment by polishing is performed after formation, so that the cross-sectional contour data of a certain layer and the lower layer The logical sum with the cross-sectional contour data is used as the correction data of the layer, and the operation of generating the correction data is performed over all the layers. Therefore, the desired dimension can be secured by cutting off the formation step after the formation, and There is an effect that a molded article with accurate dimensions can be obtained by an easy operation.
[0113]
Claim 16 The invention of claim 14 In the invention of the present invention, the loop correction process according to the surface treatment method after the formation corresponds to the case where the surface treatment by coating is performed after the formation, and the cross-sectional contour data of a certain layer and the layer one layer below the layer The logical product of the cross-sectional contour data and the correction data of the layer is used as the correction data of the layer, and the operation of generating the correction data is performed over all the layers. Dimensions can be ensured, and therefore, there is an effect that a molded article with accurate dimensions can be obtained by an easy operation.
[0114]
Claim 17 The invention of claim 14 In the invention, in the data correction process for the excess curing of the photocurable resin, the dimensional deviation of the excess curing that varies depending on the condition is preliminarily conditioned and registered in a database, and the dimensional deviation in the correction target layer at the time of the correction. From the database according to the conditions of the thickness and the number of layers of the layer to be corrected and the layer present thereabove, and processing the values searched for, and obtaining the layer below by the amount corresponding to this dimensional deviation. The logical product of the cross-sectional contour data of the layer and the correction target layer is used as the correction data of the correction target layer, and the operation of obtaining the correction data is performed for all layers except the uppermost layer. Even when the thickness of one layer is not possible, the correction of the excess curing of the photo-curable resin can be realized. Even if the part is shaped like a cup, it can be corrected to an accurate shape, and when performing a correction for a certain layer, the AND operation only needs to be performed once, This has the effect of reducing the processing time.
[0115]
Claim 18 The invention of claim 17 In the invention of the above, it is possible to specify a plurality of ranges of layers to be corrected and a range of layers not to be corrected, and to perform correction as independent shape data for each group of one layer range in the layer range to be corrected. Even if the shape has a step, there is an effect that defects in the shape due to the correction can be suppressed, and a molded object with an accurate shape can be obtained.
[0116]
Claim 19 The invention of claim 14 In the invention, the data correction process for the excess curing of the photocurable resin is performed by preconditioning and registering a dimensional deviation of the excess curing that varies depending on conditions in a database, and when the correction is performed, the correction target layer is set to the corresponding layer. Is divided into a plurality of parts according to the condition of the number of layers according to the shape of the plurality of layers existing in the upper layer, the part having no upper layer is used as correction data, and the other parts are dimensional deviations in the respective parts. Is obtained by processing the value searched or retrieved from the database according to the conditions of the thickness and the total number of layers existing above the part and the part concerned, and a layer existing below by an amount corresponding to this dimensional deviation is obtained, The logical product of the cross-sectional contour data of the layer and the portion is used as the correction data of the portion, and the correction data is obtained for all the divided portions. By integrating the correction data into correction data of the correction target layer and performing the operation of obtaining the correction data over all layers, it is possible to perform data correction capable of correcting excess hardening for any shape. effective.
[Brief description of the drawings]
FIG. 1 is a flowchart showing a flow of overall processing according to an embodiment of the present invention.
FIG. 2 is a flowchart of a stage in which three-dimensional shape data is sliced to create a plurality of cross-sectional contour lines so that a forming step is not conspicuous.
FIG. 3 is a flowchart showing a process of correcting slice coordinate values in the stage of FIG. 2;
FIG. 4 is a sectional view of a shape for explaining a correction process of FIG. 2;
FIG. 5 is a flowchart illustrating a thirassing method in which a step is not noticeable in the stage of FIG. 2;
FIG. 6 is an example of a shape for explaining the slicing method of FIG. 5;
FIG. 7 is an explanatory diagram of a facet state in FIG. 5;
FIG. 8 is an example of a shape for explaining another example of a slicing method in which steps are not noticeable in stages.
FIG. 9 is a diagram illustrating a defect example for explaining a step of correcting a defect of a cross-sectional contour line in FIG. 1;
FIG. 10 is a flowchart of a step of correcting a defect of a cross-sectional contour line in FIG. 1;
FIG. 11 is a flowchart of a process of converting a closed loop into a closed loop in the stage of FIG.
FIG. 12 is a flowchart of a closed-loop forming process of the closed loop itself in FIG. 11;
FIG. 13 is an explanatory diagram of a closed-loop processing according to the embodiment.
FIG. 14 is an explanatory diagram of another closed-loop processing of the above.
FIG. 15 is an explanatory diagram of another closed-loop processing of the above.
FIG. 16 is an explanatory diagram of another closed-loop forming process of the embodiment.
FIG. 17 is a flowchart of a closed loop forming process in which a plurality of closed loops in FIG. 11 are connected.
FIG. 18 is an explanatory diagram of the above closed loop processing.
FIG. 19 is an explanatory diagram of another closed-loop processing of the above.
FIG. 20 is an explanatory diagram of another closed-loop processing of the above.
FIG. 21 is a flowchart of another example of a closed-loop forming process in which a plurality of closed loops in FIG. 11 are connected.
FIG. 22 is an explanatory diagram of a closed-loop forming process of the embodiment.
FIG. 23 is an explanatory diagram of another closed-loop forming process of the embodiment.
FIG. 24 is an explanatory diagram of another closed-loop forming process of the above.
FIG. 25 is a flowchart of a process of dividing and deleting a closed loop self-intersection at the stage of FIG. 10;
FIG. 26 is an explanatory diagram of a process of dividing and deleting a self-intersection of the above closed loop.
FIG. 27 is a flowchart of a closed loop overlap correction process at the stage of FIG. 10;
FIG. 28 is an explanatory diagram of correction of overlapping of the closed loop according to the first embodiment.
FIG. 29 is an explanatory diagram of another example of the correction of the overlap of the above closed loop.
FIG. 30 is a flowchart of a process of deleting an unnecessary closed loop in the stage of FIG. 10;
FIG. 31 is an explanatory diagram of the above processing.
FIG. 32 is a flowchart of a process for correcting the direction of rotation of the closed loop in the stage of FIG. 10;
FIG. 33 is a flowchart of a process of correcting cross-sectional contour data in order to improve formation dimension accuracy in FIG. 1;
34 is a flowchart of data correction according to the method of surface treatment after formation in FIG. 33.
FIG. 35 is an explanatory diagram of the above processing.
36 is a flowchart of another example of data correction according to the method of surface treatment after formation in FIG. 33.
FIG. 37 is an explanatory diagram of the above processing.
FIG. 38 is a flowchart of data correction for correcting excessive curing of the photocurable resin in FIG. 33.
FIG. 39 is an explanatory diagram of the above processing.
FIG. 40 is an explanatory diagram of the above processing.
FIG. 41 is an explanatory diagram of a problem of the data correction shown in FIG. 38;
FIG. 42 is a flowchart of another example of data correction for correcting excess curing of the photocurable resin in FIG. 33.
FIG. 43 is an explanatory diagram of the above processing.
FIG. 44 is an explanatory diagram of the above processing.
FIG. 45 is an explanatory diagram of a problem of the data correction shown in FIGS. 38 and 42.
FIG. 46 is a flowchart of another example of data correction for correcting excessive curing of the photocurable resin in FIG. 33.
FIG. 47 is an explanatory diagram of the above processing.
FIG. 48 is an explanatory diagram of the above processing.
FIG. 49 is an explanatory diagram of the above processing.
FIG. 50 is an explanatory diagram of the above processing.
FIG. 51 is an explanatory diagram of a method of dividing the shape of each layer in the above.
FIG. 52 is an explanatory diagram of a method of dividing the shape of each layer in the above.
FIG. 53 is an explanatory diagram of a method of dividing the shape of each layer in the above.
FIG. 54 is an explanatory diagram of a method for dividing the shape of each layer in the above.
FIG. 55 is an explanatory diagram of a method of dividing the shape of each layer in the above.
FIG. 56 is an explanatory diagram of a conventional example.
FIG. 57 is an explanatory diagram of another conventional example.
FIG. 58 is an explanatory diagram of another conventional example.
FIG. 59 is an explanatory diagram of a problem in correction of excess curing of a photocurable resin in a conventional example.
FIG. 60 is an explanatory diagram of a problem in correction of excessive curing of a photocurable resin in a conventional example.
FIG. 61A is a cross-sectional view of a model for explaining a problem in correcting a slice coordinate value in a conventional example.
(B) is a perspective view of a modeled object for explaining a problem in correction of a slice coordinate value in a conventional example.
FIG. 62 is an example diagram of the cross-sectional contour data shown in FIG. 61;
63 is an explanatory diagram of the trajectory of laser irradiation at the time of forming a modeled object based on the cross-sectional contour data of FIG. 62.
FIG. 64 is an explanatory diagram of a conventional slicing method with less noticeable step.
FIG. 65 is an explanatory diagram of a conventional correction method by surface treatment of a modeled object.
FIG. 66 is an explanatory diagram of a problem in a conventional method of correcting the rotation direction of a closed loop of cross-sectional contour data.
FIG. 67 is an explanatory diagram of a problem in the method for correcting the rotation direction of the closed loop of the cross-sectional contour data according to the above.
FIG. 68 is a configuration diagram of a stereolithography forming apparatus used in the present invention.
FIG. 69 is a flowchart for explaining the above operation.

Claims (19)

In a three-dimensional stereolithography data creation method used in a three-dimensional shape forming apparatus that forms a desired three-dimensional shape by stacking a plurality of thin plates or a light-cured layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects in the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data, and the step of creating a plurality of cross-sectional contour data by slicing includes correcting the coordinate values to avoid slicing the three-dimensional shape data with the defective coordinate values. process and, in order to obscure the formation step, three-dimensional optical shaping de characterized by comprising a portion of prominent formation step and a step of finely sliced than normal pitch Data creation method. The slice coordinate value correcting step is a process in which, for each numerical value of the slice coordinate value sequence calculated or specified according to the condition, an arbitrary value less than the minimum value unit of the moving amount in the stacking direction of the three-dimensional shape forming apparatus whose data is to be created. 2. The method according to claim 1, wherein the step is performed by uniformly adding or subtracting a minute numerical value . In the slicing process, the shape is sliced at a constant pitch, and whether or not a layer for interpolating between two layers formed by the slice is required is determined based on the STL data existing in the region between the two layers. 2. The three-dimensional optical shaping apparatus according to claim 1 , wherein the number of interpolation layers is determined based on the normal vector information of the facet, and the slice position is determined based on the determined value to perform additional slicing between two layers . Data creation method. 4. The three-dimensional stereolithography according to claim 3, wherein data having a downward normal vector among normal vectors of facets of STL data existing in the region between the two layers is excluded from the determination . Data creation method. In a three-dimensional stereolithography data creation method used in a three-dimensional shape forming apparatus that forms a desired three-dimensional shape by stacking a plurality of thin plates or a light-cured layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects in the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy Correcting the cross-sectional contour data.The step of correcting the defect of the cross-sectional contour data includes a process of converting an open loop into a closed loop, a process of dividing and deleting a self-intersection of the closed loop, and a process of overlapping the closed loop. And the process of removing unnecessary closed loops, and defining the closed loop rotation direction in advance for data for which it is unclear whether the closed loop rotation direction is correct. Three-dimensional stereolithography data creation method characterized by comprising a step of correcting the rotational direction of the closed loop so the side is inside the shape. At the stage of closing the open loop, the open loop tries to close itself by itself, if not closed, attempts to close the loop by connecting to another open loop, and if it cannot be connected to another open loop, 6. The method according to claim 5, wherein the data is deleted . If the open loop has a self-intersection or if the end point is on the open loop, the closed loop is closed at the intersection and the rest is deleted, and the distance between the end points is specified. If the value is equal to or less than the value, a straight line is added between the endpoints to form a closed loop. 7. The data creation method for three-dimensional stereolithography according to claim 6, wherein: The closed-loop method of connecting a plurality of open loops is based on the condition that the direction of the loop has already been set correctly. The two open loops are reconnected at the intersection so that the directions match, the shorter one is deleted, and the remaining open loops are closed looped by the open loop self-closing method according to claim 7. If the process is closed, the process for both open loops is completed, and the process proceeds to the next open loop process. If the process is not closed by trial, the data of the open loop is added to the end of the process data When,
If the distance between the end point and the start point of another open loop, or the distance between the start point and the end point of another open loop is equal to or less than the specified value, if there are multiple cases, the loop between the shortest distance is the target and between the end points. connect by adding a straight line, attempting a closed loop by way of the closed loop of the open loop made by this connection at its own open loop of claim 7, applicable if it is closed by said sample line both Assuming that the process for the open loop has been completed, the process proceeds to the process of the next open loop, and if the closed loop is not performed by trial, a process of adding the data of the open loop to the end of the processing data;
For the case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, add a perpendicular from the end point to the loop, and set the loop direction from the intersection of the perpendicular and the loop. A connection is properly made, a portion remaining by the connection is deleted, and an open loop formed by the connection and the deletion is tried to make a closed loop by an open loop self-closing method according to claim 7, and the trial makes a closed loop. If it is determined that the processing for both open loops is completed, the process proceeds to the next open loop processing, and if it is not closed by trial, the process of adding the data of the open loop to the end of the processing data and ,
The process of deleting the open loop data other than the open loop to be processed in each of the above processes, until the original existing open loop and the data of the open loop secondary generated in the above process disappear, 8. The method for creating data for three-dimensional stereolithography according to claim 6 , wherein the processing of the above steps is repeated . The method of closing a loop by connecting the plurality of open loops is performed under the condition that the direction of the loop is unclear, and when a certain two open loops intersect or have a contact point, each loop has an intersection or 8. The method for self-closing an open loop according to claim 7, wherein the long sides divided at the contact points are connected, the direction of the loop is adjusted to the longer one of the two, and the rest is deleted. The closed loop is attempted by the above, if it is closed by the trial, the process for the corresponding both open loops is terminated, and the process proceeds to the next open loop. Adding data to the end of the processed data;
If the distance between the end point and the loop between other open loops is less than or equal to the specified value, if there is more than one, connect and loop by adding a straight line between the end points for the shortest distance loop Is adjusted to the longer one of the two, and the obtained open loop is tried to be closed-loop by the method of self-closed loop of open loop according to claim 7, and if the closed loop is formed by the trial, the corresponding two loops are closed. Assuming that the process for the open loop has been completed, the process proceeds to the process of the next open loop, and if the closed loop is not performed by trial, a process of adding the data of the open loop to the end of the processing data;
For a case where the distance between the end point and the loop is less than or equal to the specified value between other open loops, a perpendicular line from the end point to the loop was added, and divided at the intersection of the perpendicular line and the loop Connect to the longer one, adjust the direction of the loop to the longer one of the two, remove the rest, and try to make the resulting open loop closed loop by the open loop self closed loop method of claim 7. If the closed loop is established by the trial, the processing for the corresponding open loops is terminated, and the process proceeds to the next open loop. If the closed loop is not established by the trial, the data of the open loop is added to the end of the processing data. The process of adding to
The process of deleting the open loop data other than the open loop to be processed in each of the above processes, until the original existing open loop and the data of the open loop secondary generated in the above process disappear, 8. The method according to claim 6, wherein the steps are repeated . In the process of dividing and deleting the self-intersection, a plurality of closed loops are finally obtained by cutting out a closed loop at each intersection of the self-intersection and forming a closed loop by connecting the remaining loops. 6. The method for creating data for three-dimensional stereolithography according to claim 5, wherein data other than only the data is deleted . The process of correcting the overlap of the closed loops is to integrate the two closed loops by taking the logical sum of the two closed loops when the rotation directions of the two closed loops are equal when the rotation directions of the loops are already set correctly. 6. The three-dimensional stereolithography data generation method according to claim 5 , wherein when the rotation direction is different, the two closed loops are integrated by subtracting a small-area closed loop from a large-area closed loop . In the process of deleting the unnecessary closed loop, when the rotation direction of the loop is already set correctly, the inclusion relations of all the closed loops in one cross section are obtained in advance, and the top-level information is obtained using the obtained inclusion relation information. The closed loop is defined as a parent loop, and the difference between the rotation directions of the parent loop and the closed loop immediately below the parent loop is checked. If the parent loop has the same rotation direction as the parent loop, the closed loop is deleted. If the closed loop of the parent loop is included, the included closed loop is raised by one level, and the difference in the rotation direction between the closed loop and the parent loop is checked. 6. The method for creating data for three-dimensional stereolithography according to claim 5, wherein the operation of deleting is performed in order from the parent loop to the lowest closed loop . In the process of correcting the rotation direction of the closed loop, the inclusion relation of all closed loops in one cross section is obtained, and the top closed loop is set as a parent loop using the obtained inclusion relation information, and the inner side of the parent loop is shaped. 6. The method for creating data for three-dimensional stereolithography according to claim 5 , wherein a rotation direction is set so as to be inward of the parent loop, and a loop rotation direction opposite to that of the parent loop is sequentially set to the lowest closed loop . In a three-dimensional stereolithography data creation method used in a three-dimensional shape forming apparatus that forms a desired three-dimensional shape by stacking a plurality of thin plates or a light-cured layer formed by irradiating light to a photocurable resin, Forming a plurality of cross-sectional contour data by slicing the data so that the steps are not noticeable, correcting the defects of the cross-sectional contour data due to inadequate input shape data, and improving the forming dimension accuracy The step of correcting the cross-sectional contour data to improve the forming dimensional accuracy includes a step of correcting a loop according to a surface treatment method after forming, and a step of surplus light-curing resin. A method for producing data for three-dimensional stereolithography, which comprises a data correction process for curing. The loop correction process according to the surface treatment method after the formation corresponds to the case where the surface treatment is performed by polishing after the formation. 15. The three-dimensional stereolithography data generation method according to claim 14 , wherein an OR operation with the data is used as correction data of the layer, and an operation of generating the correction data is performed over all layers . The loop correction process according to the surface treatment method after the formation corresponds to the case where the surface treatment by coating is performed after the formation. 15. The three-dimensional stereolithography data generation method according to claim 14 , wherein an AND operation with the data is used as correction data for the layer, and the operation of generating the correction data is performed over all layers . In the data correction process for the excess curing of the photocurable resin, the dimensional deviation of the excess curing that differs depending on the condition is previously registered in a database, and the dimensional deviation in the correction target layer is corrected at the time of correction. The values obtained by searching or searching from the database according to the conditions of the thickness and the number of layers of the layer and the layer existing thereover are obtained, and the layer existing below by an amount corresponding to this dimensional deviation is obtained. 15. The method according to claim 14, wherein the logical product of the sectional contour data with the correction target layer is used as the correction data of the correction target layer, and the operation of obtaining the correction data is performed for all layers except the top layer . Data creation method for 3D stereolithography. A plurality of ranges of layers to be corrected and a range of layers not to be corrected can be specified in a plurality, and in the range of layers to be corrected, correction is performed as independent shape data for each group of one layer range. Item 18. The three-dimensional stereolithography data creation method according to Item 17 . Data correction process for excess curing above Symbol photocurable resin, previously conditioned to dimensional deviations of different excess cured by the conditions may be registered in the database, the presence of the correction target layer on top of the layer during the correction Is divided into a plurality of parts according to the condition of the number of layers according to the shape of the plurality of layers, a part having no upper layer is used as correction data, and a dimensional deviation in each part is calculated for other parts. A value obtained by processing or searching from a database according to the conditions of the thickness and the total number of layers present above the part is obtained, and a layer existing below by an amount corresponding to the dimensional deviation is obtained, and the layer and the above the logical product of the cross-sectional contour line data of the portion as the correction data of the corresponding portion, integrates these correction data with determined for all parts, separated the correction data The correction data of the correction target layer by Rukoto, three-dimensional optical shaping data creation method of claim 14, wherein the performing over an operation for obtaining the correction data in all layers.

Images