CN-121980885-A - Composite indentation data-based indentation influence factor intelligent global iteration identification method

CN121980885ACN 121980885 ACN121980885 ACN 121980885ACN-121980885-A

Abstract

The invention relates to the field of intelligent intersection of mechanical property testing and calculation of materials, in particular to an intelligent global iteration identification method for indentation influence factors based on composite indentation data. The method comprises the steps of building a mapping database based on a material constitutive model, updating the mapping database according to the residual indentation morphology and the maximum load in the mapping database, building a forward artificial neural network, training the forward artificial neural network by using the updated mapping database, obtaining a plurality of groups of residual indentation morphology and the maximum load in an indentation experiment of a material to be tested, converting the residual indentation morphology and the maximum load into actual standardized characteristic quantities irrelevant to the size of the pressure head, building a composite error function related to the residual indentation morphology and the maximum load, performing global optimization on indentation influence factors, and outputting the indentation influence factors of the material to be tested. In this way, multisource indentation information can be fused, friction and depth change are fully considered, and the method has high calculation efficiency and strengthening capability.

Inventors

JIAO QUAN
LIU HAO
Yu Daohan
LI XIUYAN
LU KE

Assignees

辽宁材料实验室

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (9)

1. An intelligent global iteration identification method for indentation influence factors based on composite indentation data is characterized by comprising the following steps: performing finite element simulation based on a material constitutive model, and establishing a mapping database, wherein the mapping database comprises residual indentation morphology, maximum load, indentation influence factors and mapping relations; Converting the residual indentation morphology and the maximum load in the mapping database into standardized characteristic quantities irrelevant to the size of the pressure head, and updating the mapping database; constructing a forward artificial neural network, and training the forward artificial neural network by using the updated mapping database to obtain a trained forward artificial neural network; obtaining a plurality of groups of residual indentation morphology and maximum load in an indentation experiment of a material to be tested, and converting the residual indentation morphology and the maximum load into actual standardized characteristic quantities irrelevant to the size of the pressure head; And constructing a composite error function related to the residual indentation morphology and the maximum load, performing global optimization on the indentation influence factor, and outputting the indentation influence factor with the minimum composite error function as the indentation influence factor of the material to be tested.
2. The method of claim 1, wherein converting the residual indentation morphology and maximum load in the mapping database into normalized feature quantities independent of indenter size, updating the mapping database comprises: geometric standardization processing is carried out on the residual indentation morphology and the maximum load, and standardized residual indentation morphology and standardized maximum load are obtained; Replacing the residual indentation morphology and the maximum load in the mapping database with a normalized residual indentation morphology and a normalized maximum load.
3. The method of claim 2, wherein the forward artificial neural network comprises an input layer, an output layer, and at least one hidden layer, wherein the hidden layer uses a ReLU activation function and the output layer uses a linear activation function; the input of the input layer is an updated mapping database, and the output of the output layer is a standardized residual indentation morphology and a standardized maximum load.
4. The method according to claim 2, wherein the geometric normalization of the residual indentation morphology comprises: Wherein, the Depth data of the residual indentation morphology; Is the radius of the spherical pressure head; Depth data for the normalized residual indentation morphology.
5. The method of claim 2, wherein geometrically normalizing the maximum load comprises: Wherein, the Is the standardized maximum load; Is the maximum load; Is the radius of the ball press head.
6. The method of claim 1, wherein training the forward artificial neural network using the updated mapping database comprises: And dividing the updated data of the mapping database into a training set and a verification set, performing iterative optimization on the forward artificial neural network by using the training set, and stopping iteration and completing training when the loss in the verification set is no longer reduced.
7. The method of claim 1, wherein the globally optimizing the indentation impact factor comprises: initializing a population, and randomly generating a plurality of candidate solutions, wherein each candidate solution is a parameter combination comprising normalized indentation influence factors; Inputting each candidate solution into a trained forward artificial neural network, outputting the standardized feature quantity of the residual indentation morphology and the maximum load, and calculating the error value of the actual standardized feature quantity of the residual indentation morphology and the maximum load of the material to be tested in an indentation experiment by utilizing the composite error function; and (3) performing iterative optimization until the error value of the individual is smaller than a preset error threshold value, and outputting the indentation influence factor of the individual.
8. The method according to claim 1, wherein obtaining a plurality of sets of residual indentation morphologies in an indentation experiment of the material to be measured comprises: And obtaining surface elevation data of the three-dimensional morphology, uniformly extracting a plurality of radial contour lines along the circumferential direction by taking the center of the residual indentation as an origin, fusing the radial contour lines into a two-dimensional axisymmetric contour, and inhibiting measurement noise by utilizing a digital filtering technology to obtain the two-dimensional residual indentation morphology.
9. An indentation influence factor intelligent global iteration recognition device based on composite indentation data is characterized by comprising: The simulation module is used for carrying out finite element simulation based on the material constitutive model and establishing a mapping database, wherein the mapping database comprises residual indentation morphology, maximum load, indentation influence factors and mapping relations; the first conversion module is used for converting the residual indentation morphology and the maximum load in the mapping database into standardized characteristic quantities irrelevant to the size of the pressure head and updating the mapping database; The network construction module is used for constructing a forward artificial neural network, and training the forward artificial neural network by utilizing the updated mapping database to obtain a trained forward artificial neural network; the second conversion module is used for obtaining a plurality of groups of residual indentation morphology and maximum load in an indentation experiment of the material to be tested and converting the residual indentation morphology and the maximum load into actual standardized characteristic quantities irrelevant to the size of the pressure head; and the output module is used for constructing a composite error function related to the residual indentation morphology and the maximum load, globally optimizing the indentation influence factor and outputting the indentation influence factor with the minimum composite error function as the indentation influence factor of the material to be tested.

Description

Composite indentation data-based indentation influence factor intelligent global iteration identification method Technical Field The invention relates to the field of intelligent intersection of material mechanical property testing and calculation, in particular to an intelligent global iteration identification method for indentation influence factors based on composite indentation data. Background The instrumented indentation technology is widely applied in the fields of material science and engineering as a micro-damage or nondestructive testing means for mechanical properties of materials. The key is that by analyzing the mechanical response (such as load-displacement curve) or the morphology feature of the residual indentation in the indentation process, the key mechanical parameters of elastic modulus, yield strength, hardening index and the like of the material are reversely pushed. The existing method is based on inversion of single information in a load-displacement curve or residual indentation morphology in the aspect of information. In addition, most of the existing methods assume ideal smooth contact or adopt an empirical friction coefficient, but in the actual indentation process, friction between a pressure head and the surface of a sample is unavoidable, and the friction coefficient changes along with materials, surface states and lubrication conditions, so that the residual indentation morphology and load response can be obviously affected. Meanwhile, the conventional algorithm is a local optimization algorithm based on gradient. However, conventional indentation inversion methods still have a number of problems in theory and practice: (1) The limitation of a single data mode is that when only a load-displacement curve is used, a large number of 'material twins' with significantly different stress-strain curves but almost indistinguishable indentation responses exist, and the uniqueness and accuracy of parameter identification are seriously affected. (2) The friction effect and the multi-working condition adaptability are ignored, the friction coefficient is not taken into an inversion system as a variable, so that the identification result deviates from the real material behavior, the comprehensive influence of the depth change on the morphology and the load characteristics is not considered by the system, and the adaptability and generalization of the method to different test working conditions (such as different loads or press-in ratios) are limited. (3) The calculation efficiency is low, the traditional inversion method based on iterative finite element simulation needs to call numerical simulation with expensive calculation for each parameter evaluation, the time consumption is serious, and quick and high-flux identification is difficult to realize. The robustness is insufficient, namely, although the problem of low finite element calculation efficiency can be solved by introducing machine learning to construct a proxy model, the traditional local optimization algorithm based on gradient is sensitive to the initial value of parameters, is easy to sink into local optimum, and influences the robustness and reliability of an inversion result. Disclosure of Invention According to the invention, an intelligent global iteration recognition scheme for the indentation influence factors based on composite indentation data is provided. The scheme can fuse multisource indentation information, fully considers friction and depth change, and has high computing efficiency and generalization capability. In a first aspect of the invention, an intelligent global iteration identification method for indentation influence factors based on composite indentation data is provided. The method comprises the following steps: performing finite element simulation based on a material constitutive model, and establishing a mapping database, wherein the mapping database comprises residual indentation morphology, maximum load, indentation influence factors and mapping relations; Converting the residual indentation morphology and the maximum load in the mapping database into standardized characteristic quantities irrelevant to the size of the pressure head, and updating the mapping database; constructing a forward artificial neural network, and training the forward artificial neural network by using the updated mapping database to obtain a trained forward artificial neural network; obtaining a plurality of groups of residual indentation morphology and maximum load in an indentation experiment of a material to be tested, and converting the residual indentation morphology and the maximum load into actual standardized characteristic quantities irrelevant to the size of the pressure head; And constructing a composite error function related to the residual indentation morphology and the maximum load, performing global optimization on the indentation influence factor, and outputting the indentation influence factor with the m