JP4625613B2

JP4625613B2 - Finite element analysis model creation system, finite element analysis model creation program, and finite element analysis model creation method

Info

Publication number: JP4625613B2
Application number: JP2003083203A
Authority: JP
Inventors: 俊樹森藤; 理聖加藤
Original assignee: Japan Research Institute Ltd
Current assignee: Japan Research Institute Ltd
Priority date: 2003-03-25
Filing date: 2003-03-25
Publication date: 2011-02-02
Anticipated expiration: 2023-03-25
Also published as: JP2004295215A

Description

【０００１】
【発明の属する技術分野】
本発明は、有限要素法を用いた各種の構造解析等のための有限要素解析モデルの作成方法及び作成プログラム等に関し、コンピュータによる構造解析技術の分野に属する。
【０００２】
【従来の技術】
各種の構造解析に用いられる有限要素法では、まず、解析対象の構造体をメッシュ状に有限要素分割してなる解析モデルが作成され、この解析モデルの作成用として種々のプログラムが提供されているが、例えば、ある構造体の強度や剛性等の改善のために、所定の部材にリブ等の補強部材を追加した場合、従来の解析モデル作成プログラムでは、補強部材を追加した状態での有限要素分割ができなかったため、次のような不具合があった。
【０００３】
つまり、この場合、両部材についてそれぞれ有限要素分割を行った後、両部材を結合して構造体全体としての解析モデルを作成することになるが、このとき、両部材の結合部において、両部材の有限要素分割でそれぞれ生成された節点が一致せず、解析モデルが節点の不整合部分を含むことになる。この状態では、以後の解析プログラムによる解析が精度よく行われず、或いはプログラムの種類によっては解析不能となる。
【０００４】
そこで、従来においては、別個に有限要素分割を行った部材をコンピュータ上で結合した後、画面を見ながら手作業で両部材の結合部における節点の位置を修正し、不整合を解消していたのであるが、そのために、多大な労力を要することとなっていた。
【０００５】
この種の問題に関し、先行特許文献１には、はり構造やシェル構造とソリッド構造との結合構造を有する構造体において、その結合境界面上で節点の不一致が発生している場合に、構造解析時に、上記結合境界面上の節点に対し、その結合状態に関連する連続条件を算出し、その条件を用いて構造解析するようにしたものが開示されている。
【０００６】
【特許文献１】
特開２０００−３３９３９１号公報
【発明が解決しようとする課題】
この特許文献１に開示された発明によれば、複数の部材の結合面上で節点の不整合が生じているモデルについての構造解析が可能となるが、節点の不一致を所定の連続条件によって補うという方法であるから、節点の不整合を消去したモデルで解析する場合に比較して精度が低下することは免れない。また、構造解析時、即ち解析プログラムの実行時に上記手法を用いるものであるから、その機能を備えた解析プログラムを用いなければ実施できず、既存の解析プログラムが使用できないなど、汎用性に欠ける面がある。
【０００７】
そこで、本発明は、上記のように複数の部材が結合されてなる構造体について有限要素法を適用する場合に、それらの部材の結合部に節点の不整合が存在しない解析モデルの作成を可能とし、これにより、手作業による節点の調整等を要することなく、汎用の解析プログラムを用いて精度よく解析できるようにすることを課題とする。
【０００８】
【課題を解決するための手段】
上記課題を解決するため、本発明は次のように構成したことを特徴とする。
【０００９】
まず、本願の請求項１に記載の発明は、第１の面部材上に第２の面部材の辺が結合されてなる構造体に有限要素法を適用するための有限要素解析モデルの作成システムに関するものであって、上記構造体の形状に関する情報を取得する構造体情報取得手段と、該手段によって取得した構造体を構成する第１、第２の面部材を単体で有限要素分割する第１有限要素分割手段と、第１の面部材上に、第２の面部材の辺を結合したときに生じる交線を形成すると共に、該交線上に上記第１有限要素分割手段によって生成された第２の面部材の上記辺上の節点をプロットする交線形成手段と、該交線形成手段によって形成された交線に沿って第１の面部材を切開すると共に、交線上にプロットされている節点のうちの切開により切り離される部分に位置する節点を切開部両側の縁部に位置する一対の節点に分割する切開手段と、該切開手段によって切開部が形成された第１の面部材を該切開部周囲に位置する節点と上記第１有限要素分割手段による有限要素分割時に該第１の面部材の外周囲に生成された節点とを用いて有限要素分割する第２有限要素分割手段と、該第２有限要素分割手段によって有限要素分割された第１の面部材の切開部を閉じて元の交線を復元すると共に、切開部両側の縁部に分割された一対の節点を上記交線上の元の位置で合体させる切開部閉鎖手段と、該閉鎖手段によって切開部が閉鎖された第１の面部材の交線上に、上記第１有限要素分割手段によって有限要素分割された第２の面部材の上記辺を結合する結合手段とを有することを特徴とする。
【００１０】
この発明によれば、解析対象構造体を構成する第１の面部材は、第２の面部材の辺との交線で切開されて、外周と切開部周囲とに輪郭を有する形状とされ、この面部材について、第２有限要素分割手段により通常の手法で有限要素分割が行われることになる。その場合に、上記交線には、第２の面部材の辺上に既に生成されている節点がプロットされると共に、そのうちの切開により切り離される部分に位置する節点は切開部両側の縁部に位置する一対の節点に分割され、これらの節点を用いて上記有限要素分割が行われる。
【００１１】
そして、その後、上記切開部が閉じられ、第１の面部材は元の交線を有する形状に復元されるが、そのとき、切開部の両側に分割された一対の節点は交線上の元の位置で合体されるので、第１の面部材の復元された交線上には、予め有限要素分割された第２の面部材の辺上の節点と同じ位置に節点が位置することになる。したがって、この第１の面部材の交線に第２の面部材の辺を結合すれば、両部材によって構成される構造体の解析モデルとして、両部材の結合部における節点の不整合を含まないモデルが作成されることになる。
【００１３】
そして、この発明によれば、第１の面部材が予め有限要素分割されているので、該第１の面部材を切開後に再び有限要素分割するときに、例えば第１の面部材の面積が切開部の面積に比較して著しく大きい場合等に、切開部周囲の一定領域を除く領域について予め生成されている節点を用いることにより、切開後の有限要素分割に要する時間が短縮されることになる。
【００１４】
また、請求項２に記載の発明は、上記請求項１に記載のシステムにおいて、第１の面部材の形状が３次元形状の場合に、交線形成手段による交線の形成及び該交線上への節点のプロットの後、該第１の面部材を所定の平面上に投影又は展開する２次元化手段と、該手段によって２次元化された仮想面部材について切開手段、第２有限要素分割手段、及び切開部閉鎖手段による処理を終了した後、該仮想面部材を当初の３次元形状に復元させる形状復元手段とを有することを特徴とする。
【００１５】
この発明によれば、上記切開手段、第２有限要素分割手段、及び切開部閉鎖手段による第１の面部材に対する各処理が、いずれも２次元の面上で行われることになり、特に第１の面部材が複雑な曲面や凹凸面で構成されている場合等に、上記各手段の構成が簡素化されると共に、これらの手段による処理に要する時間が短縮されることになる。
【００１６】
また、請求項３に記載の発明は、上記請求項２に記載のシステムにおいて、２次元化手段により第１の面部材を平面上に投影する場合に、該第１の面部材を複数の部分に分割する面部材分割手段と、該手段によって分割された各分割面部材について、上記２次元化手段、切開手段、第２有限要素分割手段、切開部閉鎖手段及び形状復元手段による各処理の後、分割された面部材を合体させる面部材合体手段とを有することを特徴とする。
【００１７】
この発明によれば、第１の面部材が例えば半円筒面等で構成されて、一つの平面に投影すると寸法が一定方向に著しく圧縮される部分が生じる場合に、この第１の面部材が分割されて、分割された各部分について２次元化以降の処理が行われるので、分割された各部材をそれぞれ最も適した方向の平面に投影することにより、第１の面部材が上記のような形状の場合にも、部分的に寸法が著しく圧縮されることが回避されることになる。したがって、投影された仮想面部材について有限要素分割した後、元の３次元形状に復元して解析モデルを作成したときに、第１の面部材の各部が良好に有限要素分割されたモデルが得られることになる。
【００１８】
つまり、投影時に寸法が部分的に著しく圧縮されると、予め第１有限要素分割手段による有限要素分割によって生成された節点間の距離が他の部分より著しく狭い個所が発生して、２次元状態での有要素分割が良好に行われないことになる。
【００１９】
しかし、上記発明により、このような不具合が解消され、第１の面部材の全体がほぼ均等な大きさの要素に分割されるなど、有限要素分割が良好に行われることになり、以後の構造解析の精度が向上することになる。
【００２０】
なお、第１の面部材を予め有限要素分割しているので、面部材分割手段により該面部材を複数の部分に分割するときに、予め生成されているメッシュに沿って分割することになる。
【００２１】
また、請求項４に記載の発明は、上記請求項１から請求項３のいずれかに記載のシステムにおいて、第１の面部材と第２の面部材の辺との交線が第１の面部材の外周辺と交わる場合であって、その交点に第２の面部材の上記辺上の節点が存在しない場合に、第２の面部材の辺における上記交点に最も近い節点を該交点に位置するように移動させる節点移動手段が設けられていることを特徴とする。
【００２２】
この発明によれば、第１の面部材を第２の面部材の辺との交線に沿って切開したときに、その切開部が第１の面部材の外周辺に連続することになるが、その場合に、上記のように交線と第１の面部材の周辺との交点に第２の面部材の節点が存在しない場合であっても、該第２の面部材の辺上の所定の節点が上記交点に移動されるから、切開部と第１の面部材の外周辺との連続部で必ず節点が存在することになる。したがって、上記のような場合であっても、第１の面部材の切開後における有限要素分割が良好に行われることになる。
【００２３】
また、請求項５に記載の発明は、上記請求項１から請求項４のいずれかに記載のシステムにおいて、切開により第１の面部材が複数の面部材に分離される場合に、第２有限要素分割手段は、分離した各面部材ごとに有限要素分割を行うことを特徴とする。
【００２４】
この発明によれば、第１の面部材上への第２の面部材の辺の交線が該第１の面部材を横断し、または該交線が円形や角形等の閉じた線である場合には、この交線に沿って第１の面部材を切開したときに、該第１の面部材が複数の部分に分離されることになるが、この場合、分離した各部分について有限要素分割が行われるので、切開部を閉じることにより分離した部分が合体されたときに、上記各発明と同様に有限要素分割された第１の面部材が得られることになる。
【００２５】
一方、請求項６に記載の発明は、第１の面部材上に第２の面部材の辺が結合されてなる構造体に有限要素法を適用するための有限要素解析モデルの作成プログラムに関するものであって、コンピュータを、上記構造体の形状に関する情報を取得する構造体情報取得手段、該手段によって取得した構造体を構成する第１、第２の面部材を単体で有限要素分割する第１有限要素分割手段、第１の面部材上に、第２の面部材の辺を結合したときに生じる交線を形成すると共に、該交線上に上記第１有限要素分割手段によって生成された第２の面部材の上記辺上の節点をプロットする交線形成手段、該交線形成手段によって形成された交線に沿って第１の面部材を切開すると共に、交線上にプロットされている節点のうちの切開により切り離される部分に位置する節点を切開部両側の縁部に位置する一対の節点に分割する切開手段、該切開手段によって切開部が形成された第１の面部材を該切開部周囲に位置する節点と上記第１有限要素分割手段による有限要素分割時に該第１の面部材の外周囲に生成された節点とを用いて有限要素分割する第２有限要素分割手段、該第２有限要素分割手段によって有限要素分割された第１の面部材の切開部を閉じて元の交線を復元すると共に、切開部両側の縁部に分割された一対の節点を上記交線上の元の位置で合体させる切開部閉鎖手段、及び、該閉鎖手段によって切開部が閉鎖された第１の面部材の交線上に、上記第１有限要素分割手段によって有限要素分割された第２の面部材の上記辺を結合する結合手段として機能させることを特徴とする。
【００２７】
また、請求項７に記載の発明は、上記請求項６に記載のプログラムにおいて、第１の面部材の形状が３次元形状の場合に、コンピュータを、交線形成手段による交線の形成及び該交線上への節点のプロットの後、該第１の面部材を所定の平面上に投影又は展開する２次元化手段、及び、該手段によって２次元化された仮想面部材について切開手段、第２有限要素分割手段、及び切開部閉鎖手段による処理を終了した後、該仮想面部材を当初の３次元形状に復元させる形状復元手段として機能させることを特徴とする。
【００２８】
また、請求項８に記載の発明は、上記請求項７に記載のプログラムにおいて、２次元化手段により第１の面部材を平面上に投影する場合に、コンピュータを、該第１の面部材を複数の部分に分割する面部材分割手段、及び、該手段によって分割された各分割面部材について、上記２次元化手段、切開手段、第２有限要素分割手段、切開部閉鎖手段及び形状復元手段による各処理の後、分割された面部材を合体させる面部材合体手段として機能させることを特徴とする。
【００２９】
また、請求項９に記載の発明は、上記請求項６から請求項８のいずれかに記載のプログラムにおいて、第１の面部材と第２の面部材の辺との交線が第１の面部材の外周辺と交わる場合であって、その交点に第２の面部材の上記辺上の節点が存在しない場合に、コンピュータを、第２の面部材の辺における上記交点に最も近い節点を該交点に位置するように移動させる節点移動手段として機能させることを特徴とする。
【００３０】
さらに、請求項１０に記載の発明は、上記請求項６から請求項９のいずれかに記載のプログラムにおいて、切開により第１の面部材が複数の面部材に分離される場合に、コンピュータを第２有限要素分割手段として機能させるときは、分離された各面部材ごとに有限要素分割を行うように機能させることを特徴とする。
【００３１】
上記の請求項６から請求項１０に記載の解析モデル作成プログラムに関する発明は、請求項１から請求項５に記載の解析モデル作成システムに関する発明とそれぞれ対応し、これらのプログラムをコンピュータに搭載して実行することにより、請求項１から請求項５のいずれかに記載のシステムと同様の作用が得られる。
【００３２】
特に、このプログラムに関する発明によれば、コンピュータを、構造体の形状に関する情報を取得する構造体情報取得手段、及び構造体を構成する面部材を単体で有限要素分割する有限要素分割手段として機能させる部分については、既存の有限要素解析用プログラムを用いることが可能であり、新たなプログラムの構築がそれだけ容易化される利点がある。
【００３５】
【発明の実施の形態】
以下、本発明の実施の形態について説明する。なお、以下の実施の形態は、本発明に係る有限要素解析モデル作成システムについての実施の形態であるが、この実施の形態において使用するプログラムは、本発明に係る有限要素解析モデル作成プログラムの実施の形態を構成する。
【００３６】
図１は本実施の形態に係るシステムの構成を示すもので、このシステム１０はコンピュータと該コンピュータに搭載されたプログラムによって構成され、これらにより、ＣＡＤ情報取得部１１、有限要素分割部１２、解析モデル作成部１３、解析部１４、及び出力部１５が設けられている。
【００３７】
上記ＣＡＤ情報取得部１１は、汎用のＣＡＤプログラム等によって作成された構造体の形状等に関する情報を取得する機能を有し、有限要素分割部１２は、与えられた各種の部材について設定された条件に従って有限要素分割する機能を有するもので、これらの部分１１、１２は、一般的なプログラムを用いて構成されている。
【００３８】
一方、解析モデル作成部１３は、上記ＣＡＤ情報取得部１１によって取得した構造体について、上記有限要素分割部１２の機能を利用しながら、その構造体全体についての有限要素解析モデルを作成するものであり、その機能については以下に詳細に説明する。
【００３９】
さらに、解析部１４は、上記解析モデル作成部１３で作成された解析モデルについて、解析目的に応じたプログラムを用いて構造解析するものであり、出力部１５は、該解析部１４による解析結果をコンピュータのモニターに表示し或いはプリントアウトするものであって、この解析部１４及び出力部１５も、既存のプログラムによって構成されている。
【００４０】
次に、フローチャート及びモニター画面図を用いて、本システム１０の解析モデル作成動作を具体的に説明する。
【００４１】
図２のフローチャートは動作の第１の具体例を示すもので、まず、ステップＳ１で、図１に示すＣＡＤ情報取得部１１により、別途作成された解析対象構造体のＣＡＤデータを取り込む。この例では、図３に示すように、円筒面状の第１面部材３０上に、平板状の第２面部材４０の下辺４１を結合してなる構造体２０が解析対象とされており、この構造体２０及びこれを構成する第１面部材３０や第２面部材４０の寸法や形状等に関する情報が取り込まれる。
【００４２】
次に、ステップＳ２で、上記構造体２０を構成する第１面部材３０と第２面部材４０について、図１に示す有限要素分割部１２の機能により、それぞれ単体で有限要素分割し、図４に示すように、メッシュ化された第１面部材３０ａ及び第２面部材４０ａを作成する。
【００４３】
そして、ステップＳ３で、図５に示すように、メッシュ化された第２面部材４０ａの下辺４１ａを、同じくメッシュ化された第１面部材３０ａ上に、当初の位置関係で結合すると共に、ステップＳ４で、図６に示すように、上記第１面部材３０ａ上に、第２面部材４０ａの下辺４１ａとの交線３１ａを作成し、この交線３１ａ上に、図５に示すメッシュ化された第２面部材４０ａの下辺４１ａに位置する図例では６個の節点Ｐ１、Ｐ２…Ｐ６をプロットし、点Ｑ１、Ｑ２…Ｑ６を作成する。
【００４４】
次に、ステップＳ５で、この第１面部材３０ａを所定の平面Ｘに投影し、図７に示すように、交線３１ｂを有する２次元化された仮想第１面部材３０ｂを作成する。その場合に、上記投影面Ｘの方向は、第１面部材３０ａの各要素Ｅ１、Ｅ２…の法線ベクトルの平均の方向と直交する方向に設定される。
【００４５】
また、ステップＳ６で、上記仮想第１面部材３０ｂについて、先に有限要素分割した際に生成された節点のうちの周辺部に位置する節点Ｒ１、Ｒ２…Ｒｎを残して内部のメッシュ化状態を解消し、図８に示すように、外周辺３２ｃ、３３ｃ、３４ｃ及び３５ｃ上の節点Ｒ１、Ｒ２…Ｒｎと、交線３１ｃ上の節点Ｑ１、Ｑ２…Ｑ６のみを有する仮想第１面部材３０ｃを作成する。
【００４６】
そして、ステップＳ７で、上記交線３１ｃに沿って仮想第１面部材３０ｃを切開し、図９に示すように、スリット状の切開部３６ｄを有する仮想第１面部材３０ｄを作成すると共に、上記交線３１ｃ上にプロットされていた節点Ｑ１、Ｑ２…Ｑ６のうちの切開により切り離される部分に位置する節点（交線３１ｃの両端に位置する節点を除く節点）Ｑ２…Ｑ５を分割して、両側の切り離し縁部に互いに対応位置する各一対の節点（Ｑ２ａ、Ｑ２ｂ）…（Ｑ５ａ、Ｑ５ｂ）を生成する。
【００４７】
ここで、第１面部材を交線に沿って切開する方法について説明する。図１０に示すように、例えば交線Ｌに４つの節点Ｐ１〜Ｐ４が存在するものとすると、これらの節点のうち、切開により切り離される部分に位置する節点Ｐ２、Ｐ３が両側に分割されることになるが、このうち、節点Ｐ２は、その両側の節点Ｐ１、Ｐ３との間の線分のなす角θ２を２等分する直線Ｌ２に沿って両側に分割され、また節点Ｐ３は、同じくその両側の節点Ｐ２、Ｐ４との間の線分のなす角θ３を２等分する直線Ｌ３に沿って両側に分割される。その場合に、当初の節点Ｐ２、Ｐ３からの距離ｄは、交線Ｌ上の節点間の距離の平均値をＬｍとし、Ｋをスケールファクター（例えば０．００１）として、次式
ｄ＝Ｋ×Ｌｍ
で示される。そして、このようにして節点を分割した後、これらの分割により生じた節点Ｐ２ａ、Ｐ２ｂ、Ｐ３ａ、Ｐ３ｂと、交線Ｌの両端に位置する節点Ｐ１、Ｐ４とを結ぶことにより、周囲に節点を有する切開部Ｙを形成する。
【００４８】
ここで、図１０は他の寸法に比して距離ｄを著しく大きく表示したもので、実際は、切開部を閉じたときの周囲の要素の変形を抑制するため、開口部として確実に把握できる範囲で極力小さな値に設定される。
【００４９】
このようにして、内部に開口した切開部を有する第１面部材が形成されることになるが、このような切開部を有する面部材の形状は、その外周囲と切開部の周囲とを輪郭とする１つの面形状として把握することができるから、図１の有限要素分割部１２を構成する一般的な有限要素分割プログラムによって分割が可能となる。
【００５０】
そこで、ステップＳ８で、図９に示す切開部３６ｄを有する仮想第１面部材３０ｄのデータを図１に示す有限要素分割部１２に引き渡し、該第１面部材３０ｄを再び有限要素分割する。その場合に、ステップＳ２の有限要素分割で生成された外周辺上の節点Ｒ１、Ｒ２…Ｒｎと、ステップＳ４、Ｓ７で生成された切開部３６ｄの周囲の節点Ｑ１、（Ｑ２ａ、Ｑ２ｂ）…（Ｑ５ａ、Ｑ５ｂ）、Ｑ６とを今回の有限要素分割の初期条件として与え、これらの節点を用いて有限要素分割する。これにより図１１に示すように、切開部３６ｅを有する有限要素分割された仮想第１面部材３０ｅが得られる。
【００５１】
次に、ステップＳ９で、上記仮想第１面部材３０ｅにおける切開部３６ｅを閉じ、図１２に示すように、切開前の交線３１ｃと同じ交線３１ｆを有する仮想第１面部材３０ｆを作成すると共に、切開時に両側に分割した各一対の節点（Ｑ２ａ、Ｑ２ｂ）…（Ｑ５ａ、Ｑ５ｂ）を元の交線３１ｃ上の位置で合体させて、第２面部材４０ａの下辺４１ａからプロットした節点Ｑ１、Ｑ２ａ…Ｑ６を復元する。
【００５２】
そして、ステップＳ１０で、図１３に示すように、仮想第１面部材３０ｆを投影前の３次元形状の第１面部材３０ｇに復元すると共に、ステップＳ１１で、この第１面部材３０ｇ上の交線３１ｇに、第２面部材４０ａの下辺４１ａを結合する。
【００５３】
このとき、図１４に示すように、第１面部材３０ｇの交線３１ｇ上の節点Ｑ１、Ｑ２…Ｑ６と、第２面部材４０の下辺４１ａの節点Ｐ１、Ｐ２…Ｐ６とが一致し、第１面部材３０上に第２面部材４０の下辺４１を結合してなる図３の構造体２０が、両部材の交線において節点の不整合を生じることなく有限要素分割されてなる解析モデル５０が完成することになる。
【００５４】
したがって、このモデル５０を用いて、図１に示す解析部１４により所定のプログラムを用いて構造解析を行えば、精度よい解析結果が得られることになる。
【００５５】
なお、以上の実施の形態は、次のような構造体にも適用可能である。
【００５６】
まず、図１５に示す構造体６０は、前述の構造体２０と同様に、第１面部材６１の面上に第２面部材６２の辺６２ａを結合させた構成であるが、この構造体６０では、第２面部材６２の上記辺６２ａが第１面部材６１の外周辺６１ａに達している。
【００５７】
この場合、第１面部材６１を第２面部材６２との交線６１ｂに沿って切開すると、同図（ｃ）に示すように、切開部６１ｃが該第１面部材６１の外周辺６１ａに接し、或いは同図（ｄ）に示すように、切開部６１ｃ′が該第１面部材６１′の外周辺６１ａ′に開口することになるが、このような第１面部材６１、６１′についても、有限要素分割部１２に備えられた既存のプログラムによって有限要素分割が可能であるから、前述の構造体２０と全く同様にして解析モデルを作成することが可能である。
【００５８】
また、図１６に示す構造体７０は、第１面部材７１の面上に、第２面部材７２の辺７２ａが該第１面部材７１を完全に横切る状態で結合されたもので、この場合、第１面部材７１を第２面部材７２との交線７１ｂに沿って切開すると、同図（ｃ）に示すように、切開部７１ｃの両端が該第１面部材７１の両側で外周辺７１ａ、７１ａに接して、該第１面部材７１が二つの部分７１ｄ、７１ｄに分断され、或いは同図（ｄ）に示すように、切開部７１ｃ′によって二つの部分７１ｄ′、７１ｄ′に完全に切り離されることになる。
【００５９】
しかし、このような場合においても、第１面部材の２つの部分７１ｄ、７１ｄ又は７１ｄ′、７１ｄ′について、有限要素分割部１２に備えられたプログラムによってそれぞれ有限要素分割を行い、その後、切開部７１ｃ、７１ｃ′を閉鎖して両部分を合体させることにより、前述の構造体２０、６０と同様に解析モデルを作成することが可能となる。
【００６０】
さらに、図１７に示す構造体８０は、第１面部材８１上に円筒面でなる面部材８２の下辺８２ａが結合されたもので、この場合、第１面部材８１を円筒面部材８２との交線８１ａに沿って切開すると、同図（ｃ）に示すように、切開部８１ｂがリング状に形成され、第１面部材８１がその内側の部分８１ｃと外側の部分８１ｄとに分離されることになる。
【００６１】
しかし、この場合も、第１面部材の２つの部分８１ｃ、８１ｄについて、有限要素分割部１２に備えられたプログラムによってそれぞれ有限要素分割を行い、その後、切開部８１ｂを閉鎖して両部分を合体させることにより、前述の各例と同様にして解析モデルを作成することが可能となる。
【００６２】
次に、図１８にフローチャートを示す解析モデル作成動作の第２の具体例について説明する。
【００６３】
ステップＴ１からステップＴ４までは前述の例と同じであって、まず、ステップＴ１で解析対象構造体のＣＡＤデータを取り込むが、この例の構造体１２０は、図１９に示すように、半円筒状の第１面部材１３０の内側上方に平板状の第２面部材１４０の下辺１４１を結合した構成とされている。
【００６４】
そして、ステップＴ２で、上記構造体１２０を構成する第１面部材１３０と第２面部材１４０をそれぞれ単体で有限要素分割し、図２０に示すように、メッシュ化された第１面部材１３０ａ及び第２面部材１４０ａを作成すると共に、ステップＴ３で、図２１に示すように、このメッシュ化された第１面部材１３０ａ上に第２面部材１４０ａの下辺１４１ａを所定の位置関係で結合する。
【００６５】
また、ステップＴ４で、図２２に示すように、上記第１面部材１３０ａ上に第２面部材１４０ａの下辺１４１ａとの交線１３１ａを作成すると共に、この交線１３１ａ上に、図２１に示すメッシュ化された第２面部材１４０ａの下辺１４１ａに位置する８個の節点Ｐ１、Ｐ２…Ｐ８をプロットし、点Ｑ１、Ｑ２…Ｑ８を作成する。
【００６６】
次に、ステップＴ５で、上記第１面部材１３０ａを左右両側に二分する。つまり、図２２に示すように、第１面部材１３０ａの中央部に、メッシュ化によって生成された要素間の境界線に沿って母線方向の分割線１３２ａを設定し、この分割先１３２ａに沿って第１面部材１３０ａを二分して、図２３に示すように、第１、第２分割第１面部材１３３ａ、１３４ａを生成する。
【００６７】
そして、これらの分割第１面部材１３３ａ、１３４ａについて、ステップＴ６からＴ１２までの処理を実行することになるが、このうちステップＴ６、Ｔ７、及びステップＴ９からステップＴ１１までの処理は、図２に示す前述の例のステップＳ５からS１０までの処理と全く同様であり、これらの処理を両分割第１面部材１３３ａ、１３４ａについて個々に実行する。
【００６８】
つまり、ステップＴ６で、第１、第２分割第１面部材１３３ａ、１３４ａを図２３に示す所定の平面Ｘ′、Ｘ″にそれぞれ投影し、図２４に示すように、２次元化された第１、第２分割仮想第１面部材１３３ｂ、１３４ｂを作成する。その場合に、上記投影面Ｘ′、Ｘ″の方向は、各分割第１面部材１３３ａ、１３４ａごとに、各要素の法線ベクトルの平均の方向と直交する方向に設定される。
【００６９】
そして、以後、第１仮想分割第１面部材１３３ｂについてのみ説明するが、ステップＴ７では、図２５に示すように、最初に有限要素分割した際に生成された節点のうちの外周辺１３５ｃ、１３６ｃ、１３７ｃに位置する節点Ｒ１、Ｒ２…Ｒｎ、及び上記分割線１３２ｃ上に位置する節点Ｒ１′、Ｒ２′…Ｒｎ′を残して内部のメッシュ化状態を解消し、これらの節点Ｒ１、Ｒ２…Ｒｎ、Ｒ１′、Ｒ２′…Ｒｎ′と、交線１３１ｃ上の節点のうち、当該第１仮想分割第１面部材１３３ｃ上の部分に位置する４個の節点Ｑ１、Ｑ２…Ｑ４を有する仮想第１面部材１３３ｃを作成する。
【００７０】
その場合に、図２５に示すように、上記交線１３１ｃと分割線１３２ｃとの交点Ｚに交線１３１ｃ上の節点が存在しないときは、ステップＴ８で、交線１３１ｃ上で交点Ｚの両側に位置する節点のうちの該交点Ｚに近い方の節点、図例では第２仮想分割第１面部材側に位置する節点Ｑ５を交点Ｚに移動させ、分割線１３２ｃ上に新たな節点Ｑ５′を生成する。このとき、図２６に示すように、その節点Ｑ５に対応する第２面部材１４０ｃの下辺１４１ｃ上の節点Ｐ５についても同様に移動させ、新たな節点Ｐ５′を生成しておく。
【００７１】
そして、ステップＴ９からステップＴ１２を前述の例と同様に実行し、まず、ステップＴ９で、上記交線１３１ｃに沿って第１仮想分割第１面部材１３３ｃを切開して、スリット状の切開部１３８ｄを有する仮想分割第１面部材１３３ｄを作成すると共に、交線１３１ｃ上にプロットされていた節点Ｑ１、Ｑ２…Ｑ４及び分割線１３２ｃ上の交点Ｚに移動させた節点Ｑ５′のうちの切開により切り離される部分に位置する節点Ｑ２…Ｑ４を分割して、両側の切り離し縁部に互いに対応位置する各一対の節点（Ｑ２ａ、Ｑ２ｂ）…（Ｑ４ａ、Ｑ４ｂ）を生成する。
【００７２】
また、ステップＴ１０で、上記の切開部１３８ｄを有する第１仮想分割第１面部材１３３ｄのデータを図１に示す有限要素分割部１２に引き渡し、該第１面部材１３３ｄを再び有限要素分割する。その場合に、図２５に示す外周辺１３５ｃ、１３６ｃ、１３７ｃ及び分割線１３２ｃ上の節点Ｒ１、Ｒ２…Ｒｎ、Ｒ１′、Ｒ２′…Ｒｎ′、及び図２７に示す切開部１３８ｄの周囲の節点Ｑ１、（Ｑ２ａ、Ｑ２ｂ）…（Ｑ４ａ、Ｑ４ｂ）、Ｑ５′とを今回の有限要素分割の初期条件として与え、これらの節点を用いて有限要素分割する。
【００７３】
これにより，図２８に示すように、切開部１３８ｅを有する有限要素分割された第１仮想分割第１面部材１３３ｅが得られる。
【００７４】
さらに、ステップＴ１１で、上記第１仮想分割第１面部材１３３ｅにおける切開部１３８ｅを閉じ、図２９に示すように、切開前の交線１３１ｃと同じ交線１３１ｆを有する仮想第１面部材１３３ｆを作成すると共に、切開時に両側に分割した各一対の節点（Ｑ２ａ、Ｑ２ｂ）…（Ｑ４ａ、Ｑ４ｂ）を、上記交線１３１ｃ上の元の位置で合体させ、節点Ｑ２…Ｑ４を復元する。
【００７５】
また、ステップＴ１２で、図２９に示す第１仮想第１面部材１３３ｆを３次元形状の第１面部材に復元すると共に、上記ステップＴ６からステップＴ１２の処理を第２分割第１面部材１３４ａについても実行し、図３０に示すように、再メッシュ化された第１、第２分割第１面部材１３３ｇ、１３４ｇを生成する。
【００７６】
そして、ステップＴ１３で、これらの分割第１面部材１３３ｇ、１３４ｇを分割線１３２ｇを合わせることにより合体させ、図３１に示すように、分割前の第１面部材１３０ａが再メッシュ化されてなる第１面部材１３０ｂを生成し、この第１面部材１３０ｂ上の交線１３１ｂに、図２６に示す第２面部材１４０ｃの下辺１４１ｃを結合する。
【００７７】
これにより、図３２に示すように、第１面部材１３０ｂの交線１３１ｂ上の節点Ｑ１、Ｑ２…Ｑ８と、第２面部材１４０ｃの下辺１４１ｃの節点Ｐ１、Ｐ２…Ｐ８とが一致し、図１９に示す第１面部材１３０上に第２面部材１４０の下辺１４１を結合してなる構造体1２０が、両部材の交線において節点の不整合を生じることなく有限要素分割されてなる解析モデル1５０が完成することになる。
【００７８】
そして、特にこの例では、第１面部材を２次元化するときに、半円筒状の第１面部材１３０ａを分割し、各分割第１面部材１３３ａ、１３４ａをそれぞれについての最適な平面Ｘ′、Ｘ″上に投影するようにしたから、投影により節点の間隔が部分的に著しく圧縮されて、２次元化した状態での再メッシュ化が良好に行われず、また、再メッシュ化後に３次元形状に復元したときに、部分的に要素の面積が著しく大きくなるといった不具合が解消され、以後の構造解析が良好に行われることになる。
【００７９】
【発明の効果】
以上のように、本発明によれば、有限要素法の適用に際し、第１の面部材上に第２の面部材の辺が結合されてなる構造体を有限要素法を適用する場合に、予め少なくとも第２の面部材を単体で有限要素分割すると共に、第１の面部材上に第２の面部材の辺を結合したときに生じる交線を形成して、該交線上に上記第２の面部材の辺上に生成された節点をプロットし、このプロットした節点を両側に分割しながら上記交線に沿って第１の面部材を切開し、この状態で該第１の面部材を有限要素分割すると共に、その後、上記切開部を閉じて分割した節点を合体させることにより、第２の面部材との交線を復元し、この交線上に第２の面部材の辺を結合するようにしたから、両部材によって構成される構造体の解析モデルとして、両部材の結合部における節点の不整合を含まないモデルが作成されることになる。
【００８０】
したがって、当該構造体についての構造解析が、節点の調整等の面倒な作業を要することなく、容易にかつ精度よく行われることになる。また、本発明によれば、第１、第２のいずれの面部材についても、既存のプログラムを用いて有限要素分割を行うことができるから、全体としてのプログラムの開発が容易である利点がる。
【図面の簡単な説明】
【図１】本発明の実施の形態に係るシステムの構成を示すブロック図である。
【図２】本発明の実施の形態に係るプログラムの第１の動作例を示すフローチャートである。
【図３】同フローチャートの第１ステップを示すモニター画面図である。
【図４】同じく第２ステップを示すモニター画面図である。
【図５】同じく第３ステップを示すモニター画面図である。
【図６】同じく第４ステップを示すモニター画面図である。
【図７】同じく第５ステップを示すモニター画面図である。
【図８】同じく第６ステップを示すモニター画面図である。
【図９】同じく第７ステップを示すモニター画面図である。
【図１０】第７ステップによる切開方法の説明図である。
【図１１】フローチャートの第８ステップを示すモニター画面図である。
【図１２】同じく第９ステップを示すモニター画面図である。
【図１３】同じく第１０ステップを示すモニター画面図である。
【図１４】同じく第１１ステップを示すモニター画面図である。
【図１５】他の応用例の説明図である。
【図１６】同じく他の応用例の説明図である。
【図１７】同じく他の応用例の説明図である。
【図１８】本発明の実施の形態に係るプログラムの第２の動作例を示すフローチャートである。
【図１９】同フローチャートの第１ステップを示すモニター画面図である。
【図２０】同じく第２ステップを示すモニター画面図である。
【図２１】同じく第３ステップを示すモニター画面図である。
【図２２】同じく第４ステップを示すモニター画面図である。
【図２３】同じく第５ステップを示すモニター画面図である。
【図２４】同じく第６ステップを示すモニター画面図である。
【図２５】同じく第７ステップを示すモニター画面図である。
【図２６】同じく第８ステップを示すモニター画面図である。
【図２７】同じく第９ステップを示すモニター画面図である。
【図２８】同じく第１０ステップを示すモニター画面図である。
【図２９】同じく第１１ステップを示すモニター画面図である。
【図３０】同じく第１２ステップを示すモニター画面図である。
【図３１】同じく第１３ステップを示すモニター画面図である。
【図３２】同じく第１４ステップを示すモニター画面図である。
【符号の説明】
１０解析モデル作成システム
２０、６０，７０、８０、１２０構造体
３０、１３０第１の面部材
４０、１４０第２の面部材[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a finite element analysis model creation method and creation program for various structural analyzes using the finite element method, and belongs to the field of computer-based structural analysis technology.
[0002]
[Prior art]
In the finite element method used for various structural analyses, first, an analysis model is created by dividing the structure to be analyzed into finite elements in a mesh shape, and various programs are provided for creating this analysis model. However, for example, when a reinforcing member such as a rib is added to a predetermined member in order to improve the strength or rigidity of a certain structure, the conventional analysis model creation program uses a finite element with the reinforcing member added. There were the following problems because it could not be divided.
[0003]
In other words, in this case, after performing finite element division for both members, the two members are combined to create an analysis model of the entire structure. Therefore, the nodes generated by the finite element division are not matched, and the analysis model includes inconsistent portions of the nodes. In this state, the subsequent analysis by the analysis program is not performed with high accuracy, or analysis is impossible depending on the type of program.
[0004]
Therefore, in the past, after joining the members that were separately divided into finite elements on the computer, the position of the node at the joint of both members was corrected manually while looking at the screen to eliminate the mismatch. However, for that purpose, much labor has been required.
[0005]
With regard to this type of problem, prior patent document 1 discloses a structural analysis in which a joint structure of a beam structure or a shell structure and a solid structure has a node mismatch on the joint boundary surface. Sometimes, a continuous condition related to the coupling state is calculated for the nodes on the coupling boundary surface, and a structure analysis is performed using the condition.
[0006]
[Patent Document 1]
JP 2000-339391 A
[Problems to be solved by the invention]
According to the invention disclosed in Patent Document 1, it is possible to perform a structural analysis on a model in which node mismatch occurs on the coupling surfaces of a plurality of members, but the node mismatch is compensated by a predetermined continuous condition. Therefore, it is inevitable that the accuracy is lowered as compared with the case of analyzing with the model in which the inconsistency of the nodes is eliminated. In addition, since the above method is used at the time of structural analysis, that is, at the time of execution of an analysis program, it cannot be performed without using an analysis program having that function, and an existing analysis program cannot be used. There is.
[0007]
Therefore, in the present invention, when the finite element method is applied to a structure in which a plurality of members are connected as described above, it is possible to create an analysis model in which there is no node mismatch at the connecting portion of these members. As a result, it is an object of the present invention to enable accurate analysis using a general-purpose analysis program without requiring manual adjustment of the nodes.
[0008]
[Means for Solving the Problems]
In order to solve the above problems, the present invention is configured as follows.
[0009]
First, the invention described in claim 1 of the present application is a system for creating a finite element analysis model for applying the finite element method to a structure in which a side of a second surface member is coupled to a first surface member. A structure information acquisition means for acquiring information related to the shape of the structure, and a structure acquired by the means The first and second surface members alone A first finite element dividing means for dividing the finite element, and an intersection line formed when the sides of the second surface member are joined on the first surface member, and the first finite element division is formed on the intersection line. An intersection line forming means for plotting the nodes on the side of the second surface member generated by the means, and cutting the first surface member along the intersection line formed by the intersection line forming means, An incision means for dividing a node located at a portion of the nodes plotted on the line to be separated by the incision into a pair of nodes located at the edges on both sides of the incision, and an incision formed by the incision means 1 node located around the incision And nodes generated on the outer periphery of the first surface member when the finite element is divided by the first finite element dividing means, A second finite element dividing means for dividing the finite element by using the first finite element dividing means, and the incision portion of the first face member divided by the second finite element dividing means is closed to restore the original intersection line, and the incision portion An incision portion closing means for joining a pair of nodes divided at both edges at the original position on the intersection line, and the intersection line of the first surface member whose incision portion is closed by the closure means. And a coupling means for coupling the sides of the second face member divided by the finite element by the one finite element dividing means.
[0010]
According to this invention, the first surface member constituting the structure to be analyzed is incised by a line of intersection with the side of the second surface member, and has a shape having an outline on the outer periphery and the periphery of the incision part, About this surface member, a finite element division is performed by a 2nd finite element division means by a normal method. In that case, nodes that have already been generated on the side of the second surface member are plotted on the intersection line, and the nodes located in the portion that is cut off by the incision are on the edges on both sides of the incision. It is divided into a pair of positioned nodes, and the finite element division is performed using these nodes.
[0011]
Thereafter, the incision is closed, and the first surface member is restored to the shape having the original line of intersection. At that time, the pair of nodes divided on both sides of the incision are the original points on the line of intersection. Since they are merged at the position, the node is located at the same position as the node on the side of the second surface member divided in advance by the finite element on the restored intersection line of the first surface member. Therefore, if the side of the second surface member is coupled to the intersection line of the first surface member, the analysis model of the structure constituted by both members does not include node mismatch at the coupling portion of both members. A model will be created.
[0013]
And According to this invention , The first surface member is a finite element in advance Because it is divided The first face member is again divided into finite elements after incision. sometimes, For example, when the area of the first surface member is remarkably larger than the area of the incised part, the finite element division after the incision is performed by using the nodes generated in advance for the area excluding the constant area around the incised part. The time required for this will be shortened.
[0014]
Also, Claim 2 The invention described in the above Claim 1 When the shape of the first surface member is a three-dimensional shape, the first surface member is defined after the formation of the intersection line by the intersection line forming means and the plotting of the nodes on the intersection line. Two-dimensionalization means for projecting or expanding on the plane of the image, and after the processing by the incision means, the second finite element dividing means, and the incision portion closing means for the virtual surface member two-dimensionalized by the means is completed, And a shape restoring means for restoring the surface member to the original three-dimensional shape.
[0015]
According to this invention, each process for the first surface member by the incision means, the second finite element dividing means, and the incision portion closing means is performed on a two-dimensional surface. When the surface member is formed of a complicated curved surface or an uneven surface, the configuration of each means is simplified and the time required for processing by these means is shortened.
[0016]
Also, Claim 3 The invention described in the above Claim 2 When the first surface member is projected onto a plane by the two-dimensional means, the surface member dividing means for dividing the first surface member into a plurality of parts, and the means divided by the means After each processing by the two-dimensionalization means, the incision means, the second finite element division means, the incision portion closing means, and the shape restoration means, for each divided surface member, a surface member combining means for combining the divided surface members. It is characterized by having.
[0017]
According to the present invention, when the first surface member is formed of, for example, a semi-cylindrical surface and a portion whose size is significantly compressed in a certain direction when projected onto one plane is generated, the first surface member is Since the divided parts are processed after the two-dimensionalization, the first surface member is as described above by projecting the divided members onto the plane in the most suitable direction. Even in the case of shapes, it is avoided that the dimensions are significantly compressed in part. Therefore, when the projected virtual surface member is divided into finite elements and then restored to the original three-dimensional shape to create an analysis model, a model in which each part of the first surface member is divided into finite elements is obtained. Will be.
[0018]
That is, if the dimensions are partly significantly compressed during projection, Generated by finite element division by the first finite element division means There are places where the distance between the nodes is significantly narrower than the other parts, and the elemental division in the two-dimensional state is performed well. Absent It will be.
[0019]
However, according to the above-described invention, such a problem is solved, and the first surface member is divided into substantially equal sized elements, so that the finite element division is favorably performed. The accuracy of analysis will be improved.
[0020]
The first surface member is divided into finite elements in advance. Because When the surface member is divided into a plurality of parts by the surface member dividing means, the surface member is divided along a mesh generated in advance.
[0021]
Also, Claim 4 The invention described in claim 1 is from the above-mentioned claim 1 Claim 3 In the system according to any one of the above, the line of intersection of the first surface member and the side of the second surface member intersects the outer periphery of the first surface member, and the second surface member at the intersection When there is no node on the side, a node moving means for moving the node closest to the intersection on the side of the second surface member so as to be located at the intersection is provided.
[0022]
According to this invention, when the first surface member is incised along the line of intersection with the side of the second surface member, the incised portion is continuous with the outer periphery of the first surface member. In this case, even when the node of the second surface member does not exist at the intersection of the intersection line and the periphery of the first surface member as described above, the predetermined value on the side of the second surface member Since the node is moved to the intersection point, the node always exists at the continuous portion between the incision portion and the outer periphery of the first surface member. Therefore, even in the above case, the finite element division after the incision of the first surface member is favorably performed.
[0023]
Also, Claim 5 The invention described in claim 1 is from the above-mentioned claim 1 Claim 4 In the system according to any one of the above, when the first surface member is separated into a plurality of surface members by incision, the second finite element dividing means performs finite element division for each separated surface member. Features.
[0024]
According to this invention, the line of intersection of the sides of the second surface member on the first surface member crosses the first surface member, or the line of intersection is a closed line such as a circle or a square. In this case, when the first surface member is cut along the intersecting line, the first surface member is separated into a plurality of parts. In this case, a finite element is used for each separated part. Since the division is performed, when the portions separated by closing the incision portion are combined, the first surface member divided into finite elements is obtained in the same manner as in the above inventions.
[0025]
on the other hand, Claim 6 The present invention relates to a computer program for creating a finite element analysis model for applying the finite element method to a structure in which a side of a second surface member is coupled to a first surface member, and a computer The structure information acquisition means for acquiring information related to the shape of the structure, and the first and second surface members constituting the structure acquired by the means are divided into finite elements alone. First finite element dividing means; On the first surface member, an intersecting line generated when the sides of the second surface member are joined is formed, and the second surface member generated by the first finite element dividing means is formed on the intersecting line. Crossing line forming means for plotting nodes on the side, cutting the first surface member along the crossing line formed by the crossing line forming means, and cutting off the cut of the nodes plotted on the crossing line An incision means for dividing a node located in the portion to be divided into a pair of nodes located at the edges on both sides of the incision, and a first surface member in which the incision is formed by the incision means The Nodes around the incision And nodes generated on the outer periphery of the first surface member when the finite element is divided by the first finite element dividing means, A second finite element dividing means for dividing the finite element using the first finite element dividing means, closing the incision portion of the first face member divided by the second finite element dividing means to restore the original line of intersection, Incision portion closing means for joining a pair of nodes divided into the edges of the first portion at the original position on the intersection line, and on the intersection line of the first surface member whose incision portion is closed by the closure means, It is made to function as a connection means which connects the said edge | side of the 2nd surface member divided | segmented into the finite element by 1 finite element division means.
[0027]
Also, Claim 7 The invention described in the above Claim 6 When the shape of the first surface member is a three-dimensional shape in the program described in 1), after the formation of the intersection line by the intersection line forming means and the plotting of the nodes on the intersection line, the first surface member The processing by the cutting means, the second finite element dividing means, and the cutting section closing means is completed for the two-dimensional means for projecting or expanding the member on a predetermined plane and the virtual surface member two-dimensionalized by the means. Then, the virtual surface member is made to function as a shape restoring means for restoring the original three-dimensional shape.
[0028]
Also, Claim 8 The invention described in the above Claim 7 When the first surface member is projected onto a plane by the two-dimensional means in the program described in the above, the computer is divided into a surface member dividing means for dividing the first surface member into a plurality of parts, and the means A surface member that combines the divided surface members after the respective processes by the two-dimensionalization means, the incision means, the second finite element division means, the incision portion closing means, and the shape restoration means. It functions as a coalescing means.
[0029]
Also, Claim 9 The invention described in the above Claims 6 to 8 In the program according to any one of the above, the line of intersection of the first surface member and the side of the second surface member intersects with the outer periphery of the first surface member, and the second surface member at the intersection When a node on the side does not exist, the computer is caused to function as a node moving means for moving the node closest to the intersection on the side of the second surface member so as to be located at the intersection. .
[0030]
further, Claim 10 The invention described in the above Claims 6 to 9 In the program described in any one of the above, when the computer functions as the second finite element dividing means when the first surface member is separated into a plurality of surface members by incision, for each separated surface member It is characterized by functioning to perform finite element division.
[0031]
above Claims 6 to 10 The invention relating to the analysis model creation program according to claim 1 starts from claim 1. Claim 5 The invention relating to the analysis model creation system according to claim 1 corresponds to the invention, and the programs are installed in a computer and executed. Claim 5 The same operation as the system described in any of the above can be obtained.
[0032]
In particular, according to the invention relating to the program, the computer is caused to function as a structure information acquisition unit that acquires information on the shape of the structure, and a finite element division unit that divides the surface member constituting the structure into a single finite element. For the portion, an existing finite element analysis program can be used, and there is an advantage that construction of a new program is facilitated accordingly.
[0035]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below. In addition, although the following embodiment is embodiment about the finite element analysis model creation system which concerns on this invention, it uses in this embodiment program Is a finite element analysis model creation according to the present invention program The embodiment is configured.
[0036]
FIG. 1 shows the configuration of a system according to the present embodiment. This system 10 is composed of a computer and a program installed in the computer. By these, a CAD information acquisition unit 11, a finite element division unit 12, an analysis A model creation unit 13, an analysis unit 14, and an output unit 15 are provided.
[0037]
The CAD information acquisition unit 11 has a function of acquiring information on the shape of the structure created by a general-purpose CAD program or the like, and the finite element division unit 12 is a condition set for various given members. The parts 11 and 12 are configured using a general program.
[0038]
On the other hand, the analysis model creation unit 13 creates a finite element analysis model for the entire structure while using the function of the finite element division unit 12 for the structure acquired by the CAD information acquisition unit 11. Yes, its function is described in detail below.
[0039]
Furthermore, the analysis unit 14 analyzes the structure of the analysis model created by the analysis model creation unit 13 using a program according to the analysis purpose, and the output unit 15 outputs the analysis result by the analysis unit 14. The data is displayed on a computer monitor or printed out, and the analysis unit 14 and the output unit 15 are also configured by existing programs.
[0040]
Next, the analysis model creation operation of the system 10 will be specifically described with reference to a flowchart and a monitor screen diagram.
[0041]
The flowchart of FIG. 2 shows a first specific example of the operation. First, in step S1, the CAD information acquisition unit 11 shown in FIG. 1 captures CAD data of a separately created analysis target structure. In this example, as shown in FIG. 3, the structure 20 formed by joining the lower side 41 of the flat plate-like second surface member 40 on the cylindrical surface-shaped first surface member 30 is an analysis target. Information relating to the size and shape of the structure 20 and the first surface member 30 and the second surface member 40 constituting the structure 20 is captured.
[0042]
Next, in step S2, the first surface member 30 and the second surface member 40 constituting the structure 20 are each finitely divided into single elements by the function of the finite element dividing unit 12 shown in FIG. As shown in FIG. 3, the meshed first surface member 30a and second surface member 40a are created.
[0043]
Then, in step S3, as shown in FIG. 5, the lower side 41a of the meshed second surface member 40a is joined to the meshed first surface member 30a in the initial positional relationship, and step In S4, as shown in FIG. 6, an intersection line 31a with the lower side 41a of the second surface member 40a is created on the first surface member 30a, and meshed as shown in FIG. 5 is formed on the intersection line 31a. In the illustrated example located on the lower side 41a of the second surface member 40a, six nodes P1, P2,... P6 are plotted, and points Q1, Q2,.
[0044]
Next, in step S5, the first surface member 30a is projected onto a predetermined plane X to create a two-dimensional virtual first surface member 30b having an intersection line 31b as shown in FIG. In that case, the direction of the projection plane X is set to a direction orthogonal to the average direction of the normal vectors of the elements E1, E2,... Of the first surface member 30a.
[0045]
Further, in step S6, the virtual first surface member 30b is subjected to the internal meshing state while leaving the nodes R1, R2,... Rn located in the periphery of the nodes generated when the finite element is divided previously. As shown in FIG. 8, the virtual first surface member 30c having only the nodes R1, R2,... Rn on the outer peripheries 32c, 33c, 34c, and 35c and the nodes Q1, Q2,. create.
[0046]
In step S7, the virtual first surface member 30c is incised along the intersection line 31c to create a virtual first surface member 30d having a slit-shaped incision 36d as shown in FIG. Nodes Q1, Q2,..., Q6 plotted on the intersection line 31c are located at the part cut off by the incision (nodes excluding the nodes located at both ends of the intersection line 31c) Q2. A pair of nodes (Q2a, Q2b)... (Q5a, Q5b) that correspond to each other at the separation edge are generated.
[0047]
Here, a method for incising the first surface member along the intersection line will be described. As shown in FIG. 10, for example, assuming that there are four nodes P1 to P4 on the intersection line L, among these nodes, the nodes P2 and P3 located at the portion separated by the incision are divided into both sides. Among these, the node P2 is divided on both sides along a straight line L2 that bisects the angle θ2 formed by the line segments between the nodes P1 and P3 on both sides thereof, and the node P3 is also divided into It is divided into both sides along a straight line L3 that bisects an angle θ3 formed by a line segment between the nodes P2 and P4 on both sides. In this case, the distance d from the initial nodes P2 and P3 is expressed by the following equation, where Lm is the average value of the distances between nodes on the intersection line L, and K is a scale factor (for example, 0.001).
d = K × Lm
Indicated by After dividing the nodes in this way, the nodes P2a, P2b, P3a, and P3b generated by these divisions are connected to the nodes P1 and P4 located at both ends of the intersection line L, so that the nodes are surrounded. An incision Y is formed.
[0048]
Here, FIG. 10 shows the distance d remarkably larger than other dimensions. Actually, in order to suppress deformation of surrounding elements when the incision is closed, a range that can be reliably grasped as an opening. The value is set as small as possible.
[0049]
In this way, the first surface member having an incision portion opened inside is formed. The shape of the surface member having such an incision portion outlines the outer periphery and the periphery of the incision portion. Therefore, the division can be performed by a general finite element division program constituting the finite element division unit 12 of FIG.
[0050]
Therefore, in step S8, the data of the virtual first surface member 30d having the incision portion 36d shown in FIG. 9 is transferred to the finite element dividing unit 12 shown in FIG. 1, and the first surface member 30d is again divided into finite elements. In that case, the nodes R1, R2,... Rn on the outer periphery generated by the finite element division in step S2 and the nodes Q1, (Q2a, Q2b) around the incision 36d generated in steps S4, S7 ( Q5a, Q5b), and Q6 are given as initial conditions for the current finite element division, and finite element division is performed using these nodes. As a result, as shown in FIG. 11, a finite element divided virtual first surface member 30e having an incision 36e is obtained.
[0051]
Next, in step S9, the incision portion 36e in the virtual first surface member 30e is closed, and as shown in FIG. 12, a virtual first surface member 30f having the same intersection line 31f as the intersection line 31c before the incision is created. In addition, the pair of nodes (Q2a, Q2b)... (Q5a, Q5b) divided on both sides at the time of incision are merged at the position on the original intersection line 31c and plotted from the lower side 41a of the second surface member 40a. , Q2a... Q6 are restored.
[0052]
Then, in step S10, as shown in FIG. 13, the virtual first surface member 30f is restored to the three-dimensional first surface member 30g before projection, and in step S11, the intersection on the first surface member 30g is restored. The lower side 41a of the second surface member 40a is coupled to the line 31g.
[0053]
At this time, as shown in FIG. 14, the nodes Q1, Q2,... Q6 on the intersection line 31g of the first surface member 30g coincide with the nodes P1, P2,. An analysis model 50 in which the structure 20 in FIG. 3 formed by connecting the lower side 41 of the second surface member 40 on the first surface member 30 is divided into finite elements without causing node mismatch at the intersection of the two members. Will be completed.
[0054]
Therefore, if the structure analysis is performed using the model 50 by the analysis unit 14 shown in FIG. 1 using a predetermined program, an accurate analysis result can be obtained.
[0055]
In addition, the above embodiment is applicable also to the following structures.
[0056]
First, the structure 60 shown in FIG. 15 has a configuration in which the side 62a of the second surface member 62 is coupled to the surface of the first surface member 61 in the same manner as the structure 20 described above. Then, the side 62 a of the second surface member 62 reaches the outer periphery 61 a of the first surface member 61.
[0057]
In this case, when the first surface member 61 is incised along the line of intersection 61 b with the second surface member 62, the incised portion 61 c is formed on the outer periphery 61 a of the first surface member 61 as shown in FIG. As shown in FIG. 4D, the incision 61c 'opens to the outer periphery 61a' of the first surface member 61 '. Such first surface members 61 and 61' In addition, since the finite element division can be performed by the existing program provided in the finite element division unit 12, it is possible to create an analysis model in exactly the same manner as the structure 20 described above.
[0058]
In addition, the structure 70 shown in FIG. 16 is formed by joining the surface 72a of the second surface member 72 on the surface of the first surface member 71 so as to completely cross the first surface member 71. When the first surface member 71 is incised along the intersecting line 71b with the second surface member 72, both ends of the incised portion 71c are on the outer periphery on both sides of the first surface member 71 as shown in FIG. The first surface member 71 is divided into two parts 71d and 71d in contact with 71a and 71a, or, as shown in FIG. 5D, the two parts 71d 'and 71d' are completely formed by an incision 71c '. Will be separated.
[0059]
However, even in such a case, the two parts 71d, 71d or 71d ', 71d' of the first surface member are each subjected to finite element division by the program provided in the finite element division unit 12, and then the incision part By closing 71c and 71c ′ and combining the two parts, an analysis model can be created in the same manner as in the structures 20 and 60 described above.
[0060]
Furthermore, the structure 80 shown in FIG. 17 is obtained by connecting a lower side 82a of a cylindrical surface member 82 on a first surface member 81. In this case, the first surface member 81 is connected to the cylindrical surface member 82. When an incision is made along the intersection line 81a, as shown in FIG. 5C, an incision 81b is formed in a ring shape, and the first surface member 81 is separated into an inner portion 81c and an outer portion 81d. It will be.
[0061]
However, also in this case, the two parts 81c and 81d of the first surface member are each subjected to finite element division by the program provided in the finite element division part 12, and then the incision part 81b is closed to combine the two parts. By doing so, it becomes possible to create an analysis model in the same manner as in the above examples.
[0062]
Next, a second specific example of the analysis model creation operation whose flowchart is shown in FIG. 18 will be described.
[0063]
Steps T1 to T4 are the same as in the above-described example. First, CAD data of a structure to be analyzed is fetched in step T1, and the structure 120 in this example has a semi-cylindrical shape as shown in FIG. The lower side 141 of the flat plate-like second surface member 140 is coupled to the inside upper side of the first surface member 130.
[0064]
Then, in step T2, the first surface member 130 and the second surface member 140 constituting the structure 120 are each singularly divided into finite elements, and as shown in FIG. 20, the meshed first surface member 130a and While creating the 2nd surface member 140a, as shown in FIG. 21, the lower side 141a of the 2nd surface member 140a is couple | bonded by predetermined | prescribed positional relationship on this meshed 1st surface member 130a by step T3.
[0065]
In step T4, as shown in FIG. 22, an intersection line 131a with the lower side 141a of the second surface member 140a is created on the first surface member 130a, and the intersection line 131a is shown in FIG. Plot the eight nodes P1, P2,... P8 located on the lower side 141a of the meshed second surface member 140a to create points Q1, Q2,.
[0066]
Next, in Step T5, the first surface member 130a is divided into two on both the left and right sides. That is, as shown in FIG. 22, a dividing line 132a in the generatrix direction is set along the boundary line between the elements generated by meshing at the center of the first surface member 130a, and along this dividing destination 132a. As shown in FIG. 23, the first surface member 130a is divided into two, and first and second divided first surface members 133a and 134a are generated.
[0067]
And about these division | segmentation 1st surface members 133a and 134a, the process from step T6 to T12 will be performed, Among these, the process from step T6, T7 and step T9 to step T11 is shown in FIG. The processing from steps S5 to S10 in the above-described example is exactly the same, and these processing are executed individually for both divided first surface members 133a and 134a.
[0068]
That is, in step T6, the first and second divided first surface members 133a and 134a are respectively projected onto the predetermined planes X ′ and X ″ shown in FIG. 23, and the two-dimensionalized first as shown in FIG. 1 and second divided virtual first surface members 133b and 134b are created. In this case, the directions of the projection surfaces X ′ and X ″ are the normal lines of the respective elements for each of the divided first surface members 133a and 134a. It is set in a direction orthogonal to the average direction of the vectors.
[0069]
In the following, only the first virtual divided first surface member 133b will be described. However, in step T7, as shown in FIG. 25, outer peripheries 135c and 136c among the nodes generated when the finite element is first divided. .., Rn and the nodes R1 ′, R2 ′,... Rn ′ located on the dividing line 132c are left to eliminate the internal meshing state, and these nodes R1, R2,. , R1 ′, R2 ′,... Rn ′, and a first virtual node having four nodes Q1, Q2,..., Q4 located at a portion on the first virtual divided first surface member 133c among the nodes on the intersection line 131c. The surface member 133c is created.
[0070]
In this case, as shown in FIG. 25, when there are no nodes on the intersection line 131c at the intersection Z between the intersection line 131c and the dividing line 132c, in step T8, on both sides of the intersection point Z on the intersection line 131c. Of the located nodes, the node closer to the intersection Z, in the illustrated example, the node Q5 located on the second virtual division first surface member side is moved to the intersection Z, and a new node Q5 ′ is placed on the division line 132c. Generate. At this time, as shown in FIG. 26, the node P5 on the lower side 141c of the second surface member 140c corresponding to the node Q5 is similarly moved to generate a new node P5 ′.
[0071]
Then, Step T9 to Step T12 are executed in the same manner as in the above example. First, in Step T9, the first virtual divided first surface member 133c is cut along the intersecting line 131c, and a slit-like cut portion 138d is formed. The virtual divided first surface member 133d having the above is created and separated by incision of the nodes Q1, Q2... Q4 plotted on the intersection line 131c and the node Q5 ′ moved to the intersection point Z on the division line 132c. The nodes Q2... Q4 located in the portion to be divided are divided to generate a pair of nodes (Q2a, Q2b)... (Q4a, Q4b) that correspond to each other at the separation edges on both sides.
[0072]
In step T10, the data of the first virtual divided first surface member 133d having the incision portion 138d is transferred to the finite element dividing portion 12 shown in FIG. 1, and the first surface member 133d is divided into finite elements again. In that case, the nodes R1, R2,... Rn, R1 ′, R2 ′,. , (Q2a, Q2b)... (Q4a, Q4b), Q5 ′ are given as initial conditions for the current finite element division, and finite element division is performed using these nodes.
[0073]
As a result, as shown in FIG. 28, a first virtual divided first surface member 133e divided into finite elements having an incision 138e is obtained.
[0074]
Further, in step T11, the incision portion 138e in the first virtual divided first surface member 133e is closed, and as shown in FIG. 29, the virtual first surface member 133f having the same intersection line 131f as the intersection line 131c before the incision is formed. A pair of nodes (Q2a, Q2b)... (Q4a, Q4b) that are created and divided on both sides at the time of incision are merged at their original positions on the intersection line 131c to restore the nodes Q2.
[0075]
In step T12, the first virtual first surface member 133f shown in FIG. 29 is restored to a three-dimensional first surface member, and the processing from step T6 to step T12 is performed on the second divided first surface member 134a. Is also executed, and the re-meshed first and second divided first surface members 133g and 134g are generated as shown in FIG.
[0076]
In step T13, the divided first surface members 133g and 134g are merged by combining the dividing lines 132g, and the first surface member 130a before the division is re-meshed as shown in FIG. The first surface member 130b is generated, and the lower side 141c of the second surface member 140c shown in FIG. 26 is coupled to the intersecting line 131b on the first surface member 130b.
[0077]
32, the nodes Q1, Q2,... Q8 on the intersection line 131b of the first surface member 130b coincide with the nodes P1, P2,... P8 of the lower side 141c of the second surface member 140c. An analysis model in which a structure 120 formed by connecting the lower side 141 of the second surface member 140 on the first surface member 130 shown in FIG. 19 is divided into finite elements without causing node mismatch at the intersection of the two members. 150 will be completed.
[0078]
Particularly in this example, when the first surface member is two-dimensionalized, the semi-cylindrical first surface member 130a is divided, and the respective divided first surface members 133a and 134a are optimally plane X ′. , X ″ is projected onto the image, the node spacing is partially compressed by projection, and remeshing in a two-dimensional state is not performed well, and three-dimensional after remeshing. When the shape is restored, the problem that the area of the element becomes partly large is solved, and the subsequent structural analysis is performed satisfactorily.
[0079]
【The invention's effect】
As described above, according to the present invention, when the finite element method is applied, a structure in which the side of the second surface member is coupled to the first surface member is applied in advance to the finite element method. At least the second surface member is divided into finite elements as a single unit, and an intersection line formed when the sides of the second surface member are coupled to the first surface member is formed, and the second line member is formed on the intersection line. The generated nodes on the side of the surface member are plotted, the first surface member is cut along the intersection line while dividing the plotted nodes on both sides, and the first surface member is finite in this state. The element is divided, and then the incision is closed and the divided nodes are combined to restore the intersection line with the second surface member, and the side of the second surface member is joined to the intersection line. Therefore, as an analysis model of the structure composed of both members, Models that do not contain mismatch definitive node is to be created.
[0080]
Therefore, the structural analysis of the structure is easily and accurately performed without requiring troublesome work such as node adjustment. Further, according to the present invention, since the finite element division can be performed using the existing program for both the first and second surface members, there is an advantage that the development of the program as a whole is easy. .
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a system according to an embodiment of the present invention.
FIG. 2 is a flowchart showing a first operation example of a program according to the embodiment of the present invention.
FIG. 3 is a monitor screen diagram showing a first step of the flowchart.
FIG. 4 is a monitor screen diagram similarly showing a second step.
FIG. 5 is a monitor screen diagram showing the third step in the same manner.
FIG. 6 is a monitor screen diagram similarly showing a fourth step.
FIG. 7 is a monitor screen diagram showing the fifth step in the same manner.
FIG. 8 is a monitor screen view similarly showing a sixth step.
FIG. 9 is a monitor screen view showing the seventh step in the same manner.
FIG. 10 is an explanatory diagram of an incision method according to a seventh step.
FIG. 11 is a monitor screen diagram showing an eighth step of the flowchart.
FIG. 12 is a monitor screen view showing the ninth step in the same manner.
FIG. 13 is a monitor screen view similarly showing the tenth step.
FIG. 14 is a monitor screen view showing the eleventh step in the same manner.
FIG. 15 is an explanatory diagram of another application example.
FIG. 16 is also an explanatory diagram of another application example.
FIG. 17 is also an explanatory diagram of another application example.
FIG. 18 is a flowchart showing a second operation example of the program according to the embodiment of the present invention.
FIG. 19 is a monitor screen view showing a first step of the flowchart.
FIG. 20 is a monitor screen diagram showing the second step in the same manner.
FIG. 21 is a monitor screen diagram showing the third step in the same manner.
FIG. 22 is a monitor screen view showing the fourth step in the same manner.
FIG. 23 is a monitor screen diagram showing the fifth step in the same manner.
FIG. 24 is a monitor screen view showing the sixth step in the same manner.
FIG. 25 is a monitor screen view showing the seventh step in the same manner.
FIG. 26 is a monitor screen view showing the eighth step in the same manner.
FIG. 27 is a monitor screen view showing the ninth step in the same manner.
FIG. 28 is a monitor screen view showing the tenth step in the same manner.
FIG. 29 is a monitor screen view showing the eleventh step in the same manner.
FIG. 30 is a monitor screen diagram showing the 12th step in the same manner.
FIG. 31 is a monitor screen view showing the thirteenth step in the same manner.
FIG. 32 is a monitor screen view showing the 14th step in the same manner.
[Explanation of symbols]
10 Analysis model creation system
20, 60, 70, 80, 120 structure
30, 130 first face member
40, 140 Second surface member

Claims

A system for creating a finite element analysis model for applying a finite element method to a structure in which a side of a second surface member is coupled to a first surface member, and acquiring information on the shape of the structure On the first surface member, a first finite element dividing means for dividing the first and second surface members constituting the structure acquired by the means into a finite element by itself, An intersection line that forms an intersection line generated when the edges of the two surface members are joined and plots a node on the edge of the second surface member generated by the first finite element dividing means on the intersection line Incising the first surface member along the intersection formed by the forming means and the intersecting line forming means, and incising the node located at the portion of the nodes plotted on the intersection line separated by the incision A pair of nodes located at the edges on both sides And incision means for dividing, the first 1 nodes and the second of the face member positioned around the incision portion of the finite element dividing means first surface member upon the finite element division by the incisions are formed by the incision means A second finite element dividing means for dividing the finite element using the nodes generated on the outer periphery, and the incision portion of the first face member divided by the second finite element dividing means is closed to restore the original intersection An incision closing means for restoring the line and combining a pair of nodes divided at the edges on both sides of the incision at the original position on the intersection line; and a first surface on which the incision is closed by the closing means A finite element analysis model creation system comprising coupling means for coupling the sides of the second surface member divided by the first finite element dividing means on the intersection line of the members.

When the shape of the first surface member is a three-dimensional shape, the first surface member is projected onto a predetermined plane after the formation of the intersection line by the intersection line forming unit and the plotting of the nodes on the intersection line. After the two-dimensional means to be expanded and the virtual surface member two-dimensionalized by the means have been processed by the incision means, the second finite element dividing means, and the incision portion closing means, the virtual surface member is returned to the original 3D. The finite element analysis model creation system according to claim 1 , further comprising shape restoration means for restoring a dimensional shape .

When projecting the first surface member onto a plane by the two-dimensional means, the surface member dividing means for dividing the first surface member into a plurality of portions, and each divided surface member divided by the means, And a surface member combining means for combining the divided surface members after each processing by the two-dimensionalizing means, the incising means, the second finite element dividing means, the incising portion closing means, and the shape restoring means. The finite element analysis model creation system according to claim 2 .

The intersection line between the first surface member and the side of the second surface member intersects the outer periphery of the first surface member, and there is no node on the side of the second surface member at the intersection. In this case, a node moving means for moving a node closest to the intersection on the side of the second surface member so as to be located at the intersection is provided. The finite element analysis model creation system described in 1.

The second finite element dividing means performs finite element division for each separated surface member when the first surface member is separated into a plurality of surface members by incision. 5. The finite element analysis model creation system according to any one of 4 above.

A program for creating a finite element analysis model for applying a finite element method to a structure in which a side of a second face member is coupled to a first face member, the computer relating to the shape of the structure On the first surface member, a structure information acquisition means for acquiring information, a first finite element dividing means for dividing the first and second surface members constituting the structure acquired by the means into a single finite element, An intersection line formed when the sides of the second surface member are joined is formed, and the nodes on the side of the second surface member generated by the first finite element dividing means are plotted on the intersection line. Incising the first surface member along the intersection formed by the line forming means and the intersection forming means, and incising a node located at a portion separated by the incision among the nodes plotted on the intersection line Located on the edges on both sides Incision means for dividing into a pair of nodes, the first surface member in which an incision is formed by the incision means is connected to the first finite element dividing means by the first finite element dividing means and the nodes located around the incision. A second finite element dividing means for dividing the finite element using the nodes generated on the outer periphery of the surface member, and closing the incision of the first surface member divided by the second finite element dividing means The incision portion closing means for combining the pair of nodes divided at the edges on both sides of the incision portion at the original position on the intersection line, and the incision portion closed by the closing means. A finite element analysis model creation program that functions as a coupling means for coupling the sides of the second surface member divided by the first finite element dividing means on the intersection line of one surface member .

When the shape of the first surface member is a three-dimensional shape, after the computer forms the intersection line by the intersection line forming means and plots the nodes on the intersection line, the first surface member is placed on the predetermined plane. Two-dimensionalization means for projecting or expanding on the virtual surface member, and processing of the virtual surface member two-dimensionalized by the means by the incision means, the second finite element dividing means, and the incision portion closing means, and then the virtual surface member The finite element analysis model creation program according to claim 6 , wherein the finite element analysis model creation program according to claim 6 is made to function as a shape restoration unit that restores the original three-dimensional shape .

When projecting the first surface member onto a plane by the two-dimensional means, the computer divides the first surface member into a plurality of parts, and each division divided by the means About a surface member, after each process by the said two-dimensionalization means, incision means, 2nd finite element division means, incision part closing means, and shape restoration means, it is made to function as a surface member combination means which unites the divided surface members. The finite element analysis model creation program according to claim 7 .

The intersection line between the first surface member and the side of the second surface member intersects the outer periphery of the first surface member, and there is no node on the side of the second surface member at the intersection. In this case, the computer is caused to function as a node moving means for moving a node closest to the intersection point on the side of the second surface member so as to be positioned at the intersection point. finite element analysis model creation program according to any.

When the first surface member is separated into a plurality of surface members by incision, when the computer functions as the second finite element dividing means, the finite element division is performed for each separated surface member. The finite element analysis model creation program according to any one of claims 6 to 9, wherein the finite element analysis model creation program.