JP3557307B2 - Melting furnace control device - Google Patents

Description

このようなルールベースマップをもとに，演算装置６ａは得られた評価値及び炉壁温度からファジィ推論の手法を用いることにより付着量の変化を推定する。この推定値を求める作業をある定められたサンプリング時間毎に行い，ｉサンプリング時間中に得られる推定値をＲｉとすると，Ｎサンプリング時点での炉壁付着量Ｒは，Ｒｉの総和により得ることができる。
【０００９】
【数１】

この得られた総和Ｒと予め与えられた付着量の下限値ＲＬとを比較して，
ＲＬ ＜ Ｒ
で示されるような制約を制御に対して与えてやることにより，炉壁の損傷を抑えることができる。また，溶融炉において非溶融物は旋回中に炉壁に付着するものとスラグとして炉下部に流れ落ちるものの合計である。従って，被溶融物供給量，燃料使用量に応じた炉壁付着量の最大値があり，ある量以上は付着しなくなる。従って，予め実際の運転を通してその最大値を求めておき，それと炉壁付着量推定値Ｒとを比較することにより，その変化が今後，最大値付近のため付着量飽和となって変化がないか，あるいは，増加継続，減少継続等いずれの状態であるかがわかる。
【００１０】
更には，その炉壁の変化量により，溶融に使用される熱量が明らかになり，炉内温度変化の未来予測を行うことができる。このための炉内温度制御系を図５に示した。先程と同様にこのような知識（第２の関係に相当）を予めオペレータへのヒアリングあるいは学習により獲得して Ｎｏ．２メモリ５ｂに記憶しておけば，演算装置６ｂは 付着量推定値，付着量変化量推定値と，ファジィ推論の手法とを用いることにより，炉内温度変化量を推定する。この場合，過去の変化量を用いることにより，現在の炉内温度の変化量を知ることができる。
以上のようにして得られた炉内温度推定値とゲインスケジュール手法等の非線形コントローラを用いて演算装置６はさらに制御操作量を計算する（図６，図７参照）。即ち，本装置Ａでは，炉壁の状態及びプロセスデータ（炉内圧力，炉内温度，非溶融物塩基度）の変化によって大きく変化する溶融スラグの流れの状態及びその時のプロセスデータを利用して予め作成しておいたルールベースから炉壁の状態を推定し，その状態及びスラグ流の状態から炉変動予測値制御操作量を決定することができる。
その結果，炉壁の損傷を積極的に低減させることにより，炉の長寿命化，保守作業の簡略化を行いながら，溶融炉の制御精度を向上させることにより，安定溶融をも実現することができる。
【００１１】
【発明の効果】
本発明に係る溶融炉制御装置は，上記したように構成されているため，炉壁の損傷を積極的に低減させることにより，炉の長寿命化，保守作業の簡略化を行いながら，溶融炉の制御精度を向上させることにより，安定溶融をも実現することができる。
【図面の簡単な説明】
【図１】本発明の実施の形態及び実施例に係る旋回流溶融炉制御装置Ａ廻りの概略構成を示す模式図。
【図２】スラグ画像の一例を示す説明図。
【図３】演算装置の構成を示すブロック図。
【図４】付着量変化ルールベースマップの一例を示す説明図。
【図５】炉内温度制御系の構成を示すブロック図。
【図６】非線形コントローラの構成を示すブロック図。
【図７】ゲインスケジュールテーブルの一例を示す説明図。
【図８】従来の旋回溶融炉制御装置の一例Ａ_０１の概略構成を示すブロック図。
【図９】従来の旋回溶融炉制御装置の他の例Ａ_０２の概略構成を示すブロック図。
【図１０】特微量抽出手法に用いるプロジェクションデータの概念を示すグラフ。
【図１１】プロジェクションデータの突出部の構造を示す図。
【符号の説明】
Ａ…旋回溶融炉制御装置
１…旋回溶融炉
２…センサ
３…カメラ
４…画像処理装置
５ａ…Ｎｏ． １メモリ（第１の記憶手段に相当）
５ｂ…Ｎｏ． ２メモリ（第２の記憶手段に相当）
６（６ａ，６ｂ）…演算装置（第１，第２の推定手段に相当）[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a melting furnace control apparatus, and more particularly, to a melting furnace control apparatus that controls a change in a furnace state using process data of a melting furnace.
[0002]
[Prior art]
Conventionally, in controlling the combustion of a melting furnace, feedback control is performed by detecting the combustion state of the melting furnace and the state of the melting output from the melting furnace and feeding it back to a controller that controls the combustion of the melting furnace. Had been done. For example, as shown in an apparatus _A01 shown in FIG. 8, the state quantity of the melting output in or from the melting furnace 30 is detected by a plurality of detectors 31 to 34 and fed back to the controllers 35 and 36, and is previously determined. The operation amount is determined by comparing with the set target value, and the melting furnace 30 is controlled based on the operation amount.
Alternatively, the physical characteristics are extracted from the image in the molten slag state and compared with the physical characteristics of the target image to judge the current reactor condition, and the operation amount for control is determined accordingly. I was For example, as in an apparatus _A02 shown in FIG. 9, the molten state of the molten fluid output from the melting furnace 40 is two-dimensionally imaged by the fusion output imaging unit 41, and the captured image is image-processed by the image processing unit 42. Then, a physical feature value relating to the image of the molten fluid is extracted, and the physical feature value is compared with a physical feature value of a target image 44 set in advance by an image deviation calculating unit 43 to obtain a deviation. The control amount of the melting furnace 40 is calculated by the control amount calculation unit 45 based on the control amount, and the combustion control of the melting furnace 40 is performed by the control unit 46.
[0003]
[Problems to be solved by the invention]
The melting furnace control device described above has the following problems.
(1) In the swirling flow melting furnace, one of the operation purposes is to keep the furnace temperature constant, but it is difficult to measure the furnace temperature due to factors such as high temperature. For this reason, the furnace wall temperature is actually measured. However, it is difficult to achieve high control accuracy by the control using the wall wall temperature error due to the furnace wall adhesion amount and the time delay with the furnace temperature due to the furnace wall brick. Therefore, it is important to measure the furnace wall adhesion amount. However, because the inside of the furnace is hot, it is sealed, and there is a lot of dust, the furnace wall adhesion amount measuring device such as an ultrasonic measurement device is melted. It is difficult to measure the amount of deposition on the furnace wall online by always installing it in the furnace. Therefore, to predict the variation of only the process data in the furnace conditions as in the conventional apparatus A _01, it is difficult to control.
(2) In the apparatus for extracting physical features from the slag molten image as in the conventional apparatus A ₀₂ is control is performed with the goal to be close to the target image, considering the state of the oven wall In the worst case, the control may cause damage to the furnace wall.
An object of the present invention is to improve a control device of a melting furnace in order to solve such problems in the prior art, and to provide a control device of a melting furnace capable of improving control accuracy by estimating a state of a furnace wall. The purpose is.
[0004]
[Means for Solving the Problems]
In order to achieve the above object, a first aspect of the present invention is a melting furnace control device for controlling fluctuations in furnace conditions using process data of a swirling melting furnace, wherein a process data including a molten slag flow state and a furnace wall temperature is provided. First storage means for storing in advance a first relationship between the molten slag and the change in the amount of furnace wall adhesion, and the present value of the process data including the flow state of the molten slag and the furnace wall temperature . A first estimating means for estimating the current change in the amount of furnace wall adhesion of the molten slag which affects the above-mentioned fluctuations in the furnace state by applying to the relation of (1). It is configured as a control device.
According to a second aspect of the present invention, there is provided a melting furnace control device for controlling fluctuations in a furnace state using process data of a melting furnace, wherein the process data including a flow state of the molten slag and a furnace wall temperature; First storage means for pre-storing a first relationship with a change in wall adhesion amount, and applying a current value of process data including a molten slag flow state and a furnace wall temperature to the first relationship. The first estimation means for estimating the current change in the amount of molten slag on the furnace wall, and the second relationship between the change in the amount of molten slag on the furnace wall and the change in the furnace temperature are stored in advance. By applying the second storage means and the estimated value of the change in the amount of furnace wall adhesion of the present molten slag to the second relationship, the current furnace temperature, which affects the change in the furnace condition, is obtained. Comprising a second estimating means for estimating a change A melting furnace control device according to claim.
According to a third aspect of the present invention, the area and the height of a portion where the value of the approximation error data, which is the difference between each pixel value in the horizontal direction of the image of the falling molten slag and a polynomial approximation value of the pixel value, protrude. And an image processing means for obtaining an index of the flow state of the molten slag based on the position and width. The applicant (Successor: Shinko Environmental Solutions Co., Ltd.) has submitted a change of applicant name to the JPO by mail sent on December 12, 2003. In addition, on the same date (February 20, 2004), a notice of proxy was submitted to the JPO by online transmission.
[0005]
DESCRIPTION OF THE PREFERRED EMBODIMENTS AND EXAMPLE
Hereinafter, embodiments and examples of the present invention will be described with reference to the accompanying drawings to provide an understanding of the present invention. The following embodiments and examples are mere examples embodying the present invention, and do not limit the technical scope of the present invention.
Here, FIG. 1 is a schematic diagram showing a schematic configuration around a swirling flow melting furnace control device A according to an embodiment and an example of the present invention, FIG. 2 is an explanatory diagram showing an example of a slag image, and FIG. FIG. 4 is an explanatory diagram showing an example of an adhesion amount change rule base map, FIG. 5 is a block diagram showing a configuration of a furnace temperature control system, and FIG. 6 is a block diagram showing a configuration of a nonlinear controller. FIG. 7 is an explanatory diagram showing an example of a gain schedule table.
[0006]
As shown in FIG. 1, a melting furnace control apparatus A according to the first and second aspects of the present invention uses a conventional example A _{01 in} that fluctuations in furnace conditions are controlled using process data from a sensor 2 of a swirling melting furnace 1. In addition, it is the same as the other example _A02 of the conventional example in that the variation in the furnace state is controlled using image data obtained from the CCD camera 3 via the image processing device 4. However, in the present apparatus A, the first relationship between the flow state of the molten slag and the process data and the change in the amount of furnace wall adhesion of the molten slag is stored in advance. By applying the memory 5a (corresponding to a first storage means) and the flow state of the molten slag and the current value of the process data to the above-mentioned first relationship, the current molten slag furnace wall which affects the furnace state is applied. This is different from the conventional example in that a calculation device 6a (corresponding to first estimating means) for estimating a change in the amount of adhesion is provided (first invention). Further, in addition to the above elements, the present apparatus A further stores in advance a second relationship between a change in the amount of molten slag adhering to the furnace wall and a change in the furnace temperature. By applying the estimated value of the change in the amount of furnace wall adhesion of the molten slag to the second relationship, the current memory 5b (corresponding to the second storage means) and the current relationship affecting the fluctuations in the furnace state are applied. The second embodiment is different from the conventional example in that a calculation device 6b (corresponding to a second estimation unit) for estimating a change in the furnace temperature is provided (second invention).
Hereinafter, the device A will be further embodied, and its operation will be described.
[0007]
In FIG. 1, an imaging device such as a CCD camera 3 is installed below the swirling melting furnace 1 to photograph a state in which slag is dropped. The image signal obtained therefrom, for example, as shown in FIG. 2, is transmitted to an image processing device 4 which can take in the image signal. In the image processing apparatus 4, extraction of characteristic amounts (flow rate, stable amount of flow, slag width, slag speed, etc.) and calculation of evaluation values are performed as needed to create time-series data of evaluation values.
The feature amount is extracted as follows. An image (hereinafter, projection data) obtained by a CCD camera is as shown in FIG. 10. First, each pixel value is projected in the horizontal direction of the screen to calculate a sum.
Next, an approximation formula is obtained by applying a polynomial approximation technique to the projection data, and a difference between the approximation formula and the projection data is calculated to obtain approximation error data (FIG. 10).
Further, the protruding portion in the approximation error data is regarded as a slag, and the portion is extracted, and the area (AR), height (HI), position (PO), width (WD), etc. of the protruding portion are calculated. See FIG. 11 for the concept of each element.
Next, this operation is performed on N consecutive images, and the average value of AR (AR2), the average value of HI (HI2), the reciprocal of the standard deviation value of PO (PO2), and the average value of WD (WD2) Is calculated. AR2 is the slag flow rate, HI2 is the slag speed, PO2 is the flow stability, WD2 can be regarded as approximating the slag width, and the physical characteristics including the temporal variation of the slag can be considered by these features. Can be quantified. Since the flow of PO is considered to be more stable as the fluctuation of PO is smaller, the reciprocal of the standard deviation of PO is used as an index of flow stability.
As the evaluation value calculation, for example, the following can be considered.
J = W1 * (flow rate) + W2 * (stable amount of flow) + W3 * (slag width) + W4 * (slag speed)
At the same time, time-series data of process data such as the furnace wall temperature and the furnace pressure at that time is taken into the arithmetic unit 6 (6a, 6b) from a process computer (not shown) (see FIG. 3). In the following, for convenience of explanation, a case where the furnace wall temperature is used as an example of the process data will be described.
[0008]
Using the information on the process (evaluation value, furnace wall temperature) independently obtained in this way, the knowledge of the operator is extracted and created in advance, and no. One is compared with the rule base map (corresponding to the first relationship) stored in the memory 5a.
For example, a rule as shown in FIG. 4 can be considered as this rule base map. This can be obtained by hearing or learning from a skilled operator, and has the following meaning.

Based on such a rule base map, the arithmetic unit 6a estimates a change in the amount of adhesion from the obtained evaluation value and the furnace wall temperature by using a fuzzy inference technique. When the work of obtaining the estimated value is performed at every predetermined sampling time, and the estimated value obtained during the i sampling time is defined as Ri, the furnace wall adhesion amount R at the N sampling time can be obtained by the sum of Ri. it can.
[0009]
(Equation 1)

The obtained sum R is compared with a predetermined lower limit value RL of the adhesion amount, and
RL <R
By giving the constraint as shown in the above to the control, damage to the furnace wall can be suppressed. In the melting furnace, the unmelted material is the sum of the material that adheres to the furnace wall during turning and the material that flows down to the lower part of the furnace as slag. Therefore, there is a maximum value of the furnace wall adhesion amount according to the supply amount of the molten material and the fuel consumption amount, and the amount of the furnace wall adhesion does not exceed a certain amount. Therefore, the maximum value is determined in advance through actual operation, and the calculated value is compared with the estimated value R of the furnace wall adhesion. , Or whether the state is continuous increase or decrease.
[0010]
Further, the amount of heat used for melting becomes clear from the change amount of the furnace wall, and the future prediction of the furnace temperature change can be performed. FIG. 5 shows a furnace temperature control system for this purpose. As in the previous case, such knowledge (corresponding to the second relationship) is obtained in advance by interviewing or learning with the operator, and No. If it is stored in the memory 5b, the arithmetic unit 6b estimates the temperature change in the furnace by using the estimated adhesion amount, the estimated adhesion change amount, and the fuzzy inference method. In this case, the current amount of change in the furnace temperature can be known by using the amount of change in the past.
The arithmetic unit 6 further calculates a control operation amount using the furnace temperature estimated value obtained as described above and a non-linear controller such as a gain schedule method (see FIGS. 6 and 7). That is, the present apparatus A utilizes the state of the molten slag flow, which greatly changes due to changes in the state of the furnace wall and the process data (furnace pressure, furnace temperature, non-moltenity basicity), and the process data at that time. It is possible to estimate the state of the furnace wall from a rule base created in advance, and to determine the control operation amount of the furnace fluctuation predicted value from the state and the state of the slag flow.
As a result, it is possible to achieve stable melting by improving the control accuracy of the melting furnace while prolonging the life of the furnace and simplifying maintenance work by actively reducing damage to the furnace wall. it can.
[0011]
【The invention's effect】
Since the melting furnace control device according to the present invention is configured as described above, by actively reducing damage to the furnace wall, it is possible to extend the life of the furnace and simplify maintenance work while maintaining the melting furnace. By improving the control accuracy, stable melting can be realized.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing a schematic configuration around a swirling flow melting furnace control device A according to an embodiment and an example of the present invention.
FIG. 2 is an explanatory diagram showing an example of a slag image.
FIG. 3 is a block diagram illustrating a configuration of an arithmetic unit.
FIG. 4 is an explanatory diagram showing an example of an adhesion amount change rule base map.
FIG. 5 is a block diagram showing a configuration of a furnace temperature control system.
FIG. 6 is a block diagram illustrating a configuration of a nonlinear controller.
FIG. 7 is an explanatory diagram showing an example of a gain schedule table.
FIG. 8 is a block diagram showing a schematic configuration of an example _A01 of a conventional rotary melting furnace control device.
FIG. 9 is a block diagram showing a schematic configuration of another example _A02 of the conventional rotary melting furnace control device.
FIG. 10 is a graph showing the concept of projection data used for a very small amount extraction technique.
FIG. 11 is a diagram showing a structure of a projection of projection data.
[Explanation of symbols]
A: Swing melting furnace control device 1 ... Swirl melting furnace 2 ... Sensor 3 ... Camera 4 ... Image processing device 5a ... No. 1 memory (corresponding to first storage means)
5b ... No. 2 memories (corresponding to the second storage means)
6 (6a, 6b): arithmetic unit (corresponding to first and second estimating means)

Claims (3)

In a melting furnace control device that controls fluctuations in the furnace conditions using the process data of a swirling melting furnace, the process data including the flow state of the molten slag and the furnace wall temperature and the change in the amount of furnace wall adhesion of the molten slag are compared. A first storage means for storing the first relation in advance and a current value of the process data including the flow state of the molten slag and the furnace wall temperature are applied to the first relation, whereby the condition in the furnace is obtained. A first estimating means for estimating a change in the current amount of the molten slag attached to the furnace wall affecting the fluctuation, the melting furnace control device comprising: In a melting furnace control device that controls fluctuations in the furnace conditions using the process data of a swirling melting furnace, the process data including the flow state of the molten slag and the furnace wall temperature and the change in the amount of furnace wall adhesion of the molten slag are compared. A first storage means for storing the first relation in advance, and a present value of process data including a flow state of the molten slag and a furnace wall temperature are applied to the first relation to obtain a current molten slag. First estimating means for estimating a change in furnace wall adhesion amount, and second storage means for preliminarily storing a second relationship between a change in the furnace wall adhesion amount of molten slag and a change in furnace temperature. And applying the current estimated value of the change in the amount of molten slag to the furnace wall to the second relationship, thereby estimating the current change in the furnace temperature affecting the change in the furnace condition. Melting means comprising estimating means. The control device. The molten slag is determined based on the area, height, position and width of the portion where the value of the approximation error data, which is the difference between each pixel value in the horizontal direction of the image of the falling molten slag and a polynomial approximation of the pixel value, protrudes. 3. The melting furnace control device according to claim 1, further comprising image processing means for obtaining an index of the flow state of the melting furnace.

Images