CN-117034039-B - Method and system for determining similarity of casting solidification process

CN117034039BCN 117034039 BCN117034039 BCN 117034039BCN-117034039-B

Abstract

A method and a system for determining similarity of casting solidification process relate to the technical field of casting solidification. The invention aims to solve the problem of low accuracy of the existing method for determining the similarity of the casting solidification process. The method comprises the steps of obtaining the modulus, pouring temperature, sand mold temperature and heat exchange coefficient of two castings to be evaluated for solidification similarity, respectively inputting the modulus, pouring temperature, sand mold temperature and heat exchange coefficient of the two castings to be evaluated for solidification similarity into an intelligent casting analysis model to obtain niyama criterion values of the two castings to be evaluated for solidification similarity, and obtaining a solidification similarity conclusion of the two castings to be evaluated for solidification similarity by utilizing the niyama criterion values. The method is used for determining whether the solidification process of the castings is similar.

Inventors

HUANG XIXI
XUE XIANG
WU SHIPING
ZHU JIHU
WANG MINGJIE
DAI GUIXIN

Assignees

哈尔滨工业大学

Dates

Publication Date: 20260512
Application Date: 20230911

Claims (10)

1. A method for determining the similarity of casting solidification process is characterized by comprising the following specific steps: S1, obtaining the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of two castings of which the solidification similarity is to be evaluated, and respectively inputting the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of the two castings of which the solidification similarity is to be evaluated into a trained intelligent casting analysis model to obtain niyama criterion values of the two castings of which the solidification similarity is to be evaluated; S2, obtaining a solidification similarity conclusion of two castings to be evaluated for solidification similarity by utilizing the niyama criterion value obtained in the S1; the trained intelligent casting analysis model is obtained by the following steps: step one, a casting solidification model is established, and a casting solidification process is simulated, so that a casting solidification process simulation result is obtained; Step two, acquiring casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient from a casting solidification process simulation result, and acquiring the casting solidification model under the current casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient A point(s), A point(s), A temperature gradient G and a cooling rate L of the spot, thereby obtaining A point(s), A point(s), Niyama criteria of points; in casting solidification models A point(s), A point(s), The points are as follows: Obtaining a vertex of a solidification model of a casting And the current vertex P 2 (0,L/2,L/2)、P 3 (L/2 ) is set as the origin of the space rectangular coordinate system; step three, casting modulus, pouring temperature, sand mold temperature, heat exchange coefficient and casting temperature obtained in the step two A point(s), A point(s), The niyama criteria of the points form a training set, the intelligent casting analysis model is trained by the training set, and the intelligent casting analysis model is trained.
2. A method for determining the similarity of solidification processes of castings according to claim 1, wherein said method is characterized in that A point(s), A point(s), The niyama criteria for a point are obtained by the following formula: where G is the temperature gradient and L is the cooling rate.
3. The method for determining the similarity of casting solidification process according to claim 2, wherein the intelligent casting analysis model is a three-layer artificial neural network and comprises an input layer, a hidden layer and an output layer; The hidden layers are two layers.
4. A method for determining the similarity of solidification processes of castings according to claim 3, wherein the loss function of the intelligent casting analysis model is represented by the following formula: Wherein, the Is the value of the output of the target, Is a model predictive value of the model, and, , i Is a regularization parameter.
5. The method for determining the solidification process similarity of castings according to claim 4, wherein the niyama criterion value obtained by using S1 in S2 is used for obtaining a solidification similarity conclusion of two castings to be evaluated for solidification similarity, and the method is specifically as follows: If on two castings to be evaluated for solidification similarity A point(s), A point(s), The niyama criteria of the points are all standard-identical, so that the solidification processes of the two castings with the solidification similarity to be evaluated are similar, and if the niyama criteria of any point are not standard-identical, the solidification processes of the two castings with the solidification similarity to be evaluated are dissimilar; The standard equality is that the errors of the two niyama criteria are within the preset error.
6. A casting solidification process similarity determination system for executing a casting solidification process similarity determination method according to any one of claims 1 to 5, wherein the system comprises a casting parameter acquisition module, a niyama criterion value acquisition module and a solidification similarity determination module; The casting parameter acquisition module is used for acquiring the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of the two castings of which the solidification similarity is to be evaluated, and sending the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of the two castings of which the solidification similarity is to be evaluated to the niyama criterion value acquisition module; The niyama criterion value acquisition module is used for respectively inputting the modulus, pouring temperature, sand mold temperature and heat exchange coefficient of each casting to be evaluated for solidification similarity into the trained intelligent casting analysis model to obtain niyama criterion values of each casting to be evaluated for solidification similarity, and inputting niyama criterion values of each casting to be evaluated for solidification similarity into the solidification similarity determination module; and the solidification similarity determining module is used for obtaining solidification similarity conclusions of the two castings of which the solidification similarities are to be evaluated by utilizing niyama judgment values of each casting of which the solidification similarities are to be evaluated.
7. The casting solidification process similarity determination system of claim 6, wherein the solidification similarity analysis software is obtained by: step1, establishing a casting solidification model, and simulating a casting solidification process to obtain a casting solidification process simulation result; Step2, casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient in the casting solidification process simulation result, and obtaining the casting solidification model under the current casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient A point(s), A point(s), A temperature gradient G and a cooling rate L of the spot, thereby obtaining A point(s), A point(s), Niyama criterion values of points; in casting solidification models A point(s), A point(s), The points are as follows: Obtaining a vertex of a solidification model of a casting And the current vertex Setting P 2 (0,L/2,L/2)、P 3 (L/2 ) as an origin, constructing an x axis and a y axis along the side length direction of the casting; niyama criteria value obtained by the following formula: step3, casting modulus, pouring temperature, sand mold temperature, heat exchange coefficient and casting temperature obtained by Step2 A point(s), A point(s), And the niyama criteria of the points form a training set, and the intelligent casting analysis model is trained by utilizing the training set to obtain a trained intelligent casting analysis model.
8. The casting solidification process similarity determination system of claim 7, wherein the intelligent casting analysis model is a three-layer artificial neural network comprising an input layer, a hidden layer and an output layer; The hidden layers are two layers.
9. The casting solidification process similarity determination system of claim 8, wherein the intelligent casting analysis model has a loss function of the following formula: Wherein, the Is the value of the output of the target, Is a model predictive value of the model, and, , Is a regularization parameter.
10. The system for determining the solidification process similarity of castings according to claim 9, wherein the method is characterized in that by utilizing niyama judgment values of each casting with solidification similarity to be evaluated, solidification similarity conclusions of two castings with solidification similarity to be evaluated are obtained, and specifically: If on two castings to be evaluated for solidification similarity A point(s), A point(s), The niyama criteria of the points are all standard-identical, so that the solidification processes of the two castings with the solidification similarity to be evaluated are similar, and if the niyama criteria of any point are not standard-identical, the solidification processes of the two castings with the solidification similarity to be evaluated are dissimilar; The standard equality is that the errors of the two niyama criteria are within the preset error.

Description

Method and system for determining similarity of casting solidification process Technical Field The invention relates to the technical field of casting solidification, in particular to a method and a system for determining similarity of casting solidification processes. Background In order to analyze the defect formation rule of the casting, the solidification process of the part with the solidification defect of the casting is required to be researched by reducing or enlarging the size of the solidification process of the part with the solidification defect of the casting, transferring the research result to the original casting again, further guiding the production of the casting, and sometimes, the casting process of the casting obtained under the laboratory condition is required to be transferred to the actual production or a process of a part or a whole casting. All this is accomplished without a problem, i.e. similar solidification. The bridge cannot be erected between two solidification phenomena due to poor solidification similarity, so that research on the solidification similarity problem has important theoretical significance and practical application value for exploring the evolution rule of two solidification processes and obtaining a control method for the two solidification processes. At present, the similarity of two solidification processes is mainly judged by adopting a temperature similarity criterion or a flow similarity criterion, and the similarity of the solidification processes of castings is mainly judged by adopting a single criterion, such as solidification temperature gradient, solidification time, solid phase rate or cooling rate, but the accuracy is low due to the fact that the solidification processes influence factors are numerous and the similarity of the solidification processes of castings is judged by adopting only the single criterion. Disclosure of Invention The invention aims to solve the problem of low accuracy of the existing method for determining the similarity of the casting solidification process, and provides a method and a system for determining the similarity of the casting solidification process. The casting solidification process similarity determination method comprises the following specific processes: S1, obtaining the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of two castings of which the solidification similarity is to be evaluated, and respectively inputting the modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient of the two castings of which the solidification similarity is to be evaluated into a trained intelligent casting analysis model to obtain niyama criterion values of the two castings of which the solidification similarity is to be evaluated; S2, obtaining a solidification similarity conclusion of two castings to be evaluated for solidification similarity by utilizing the niyama criterion value obtained in the S1; the trained intelligent casting analysis model is obtained by the following steps: step one, a casting solidification model is established, and a casting solidification process is simulated, so that a casting solidification process simulation result is obtained; Step two, acquiring casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient in a casting solidification process simulation result, and acquiring temperature gradient G and cooling rate L of a point P 1, a point P 2 and a point P 3 on a casting solidification model under the current casting modulus, pouring temperature, sand mold temperature and heat exchange coefficient, thereby acquiring niyama criteria of a point P 1, a point P 2 and a point P 3; The points P 1, P 2 and P 3 in the casting solidification model are as follows: Obtaining a vertex P 1 of the casting solidification model, taking the current vertex P 1 as an origin of a space rectangular coordinate system, and setting P 2(0,L/2,L/2)、P3 (L/2, L/2 and L/2); And thirdly, forming a training set by the casting modulus, the pouring temperature, the sand mold temperature and the heat exchange coefficient obtained in the second step and niyama criteria of the point P 1, the point P 2 and the point P 3, and training an intelligent casting analysis model by using the training set. Further, niyama criteria of the point P 1, the point P 2 and the point P 3 are obtained by the following formulas: where G is the temperature gradient and L is the cooling rate. Further, the intelligent casting analysis model is a three-layer artificial neural network and comprises an input layer, a hidden layer and an output layer; The hidden layers are two layers. Further, the loss function of the intelligent casting analysis model is as follows: Where O is the target output value, a is the model predicted value, λ, θ i is the regularization parameter. Further, the niyama criterion value obtained by using the S1 in the S2