Method for reduced order treatment of direct coupling of multi-physics fields of transformer and related device

Abstract

The embodiment of the application discloses a transformer multi-physical field direct coupling reduced-order processing method and related device, method includes: obtaining and using the geometric parameters, material properties and operation condition of the transformer, establishing the simulation geometric model of the transformer; based on the heat generation, fluid flow and fluid-solid heat transfer process of the transformer operation, the initial coupling model of the thermal-fluid-solid multi-physical field direct coupling of the transformer is constructed; the reduced-order model of the thermal-fluid-solid multi-physical field is established by using the random forest algorithm and the initial coupling model, the multi-physical field operation simulation of the transformer is carried out according to the simulation geometric model and the reduced-order model, and the operation simulation result reflecting the operation condition of the transformer in the multi-physical field is obtained. Through the random forest algorithm for the reduction of the initial coupling model, the complexity of the model can be simplified, the degrees of freedom in the simulation model are reduced, in the multi-physical field operation simulation, the calculation efficiency of the simulation can be improved, the demand for computing resources is reduced, and the reliability of the result is maintained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transformer multi-physics field coupling calculation, and in particular to a transformer multi-physics field direct coupling reduction processing method and related device. BACKGROUND

[0002] In transformer operation and selection, multi-physics field simulation is involved for margin checking. In the operation process of the transformer, heat is generated due to copper loss, iron loss, etc., and is transferred through the medium (transformer oil, insulating material, shell, air, etc.). The flow of the transformer oil is affected by the internal structure of the transformer (such as the winding, the iron core, etc.), and at the same time, the flow of the oil also affects the thermal distribution of the entire system, thereby affecting the efficiency and safety of the transformer. In addition, the propagation of heat will cause thermal expansion of solid components, thereby affecting the mechanical structure and electrical performance of the transformer.

[0003] The above physical processes exist simultaneously and interact with each other. When directly coupling the calculation of multi-physics fields, simulation usually needs to consume huge computing resources, especially when fine grids and complex boundary conditions are involved. Due to the mutual coupling of heat conduction, fluid flow and solid deformation processes, traditional numerical simulation methods often have difficulty in balancing between calculation efficiency and accuracy when dealing with such problems.

[0004] Therefore, there is an urgent need for a means that can improve the calculation efficiency and accuracy when directly coupling the calculation of multi-physics fields. SUMMARY

[0005] The main purpose of the present application is to provide a transformer multi-physics field direct coupling reduction processing method and related device, which can solve the problem of lack of means to improve the calculation efficiency and accuracy when directly coupling the calculation of multi-physics fields in the prior art.

[0006] To achieve the above purpose, the first aspect of the present application provides a transformer multi-physics field direct coupling reduction processing method, which comprises:

[0007] Obtaining and utilizing the geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer;

[0008] Based on the heat generation, fluid flow and fluid-solid heat transfer processes of the transformer in operation, an initial coupling model of the thermal-fluid-solid multi-physics field direct coupling of the transformer is constructed;

[0009] Using a random forest and the initial coupling model, a reduction model of the thermal-fluid-solid multi-physics field is established;

[0010] The multi-physical field operation simulation of the transformer is performed according to the simulation geometric model and the reduced-order model, and operation simulation results are obtained, which reflect the operation conditions of the transformer in the multi-physical field.

[0011] In an implementation, the reduced-order model of the thermal-hydraulic-mechanical multi-physical field is established by using the random forest and the initial coupling model, and includes:

[0012] The importance of all candidate data features is calculated by the random forest, the candidate data features being used to indicate data features affecting the operation simulation results in the multi-physical field operation simulation of the transformer;

[0013] The Pearson correlation coefficients between the candidate data features are calculated;

[0014] Whether the candidate data features are strongly correlated is determined based on the Pearson correlation coefficients;

[0015] If the candidate data features are strongly correlated, the candidate data features are determined as redundant features;

[0016] The target data features with low importance in the redundant features are removed from the initial coupling model to obtain the reduced-order model.

[0017] In an implementation, the reduced-order model of the thermal-hydraulic-mechanical multi-physical field is established by using the random forest and the initial coupling model, and includes:

[0018] The importance of all candidate data features is calculated by the random forest, the candidate data features being used to indicate data features affecting the operation simulation results in the multi-physical field operation simulation of the transformer;

[0019] The Pearson correlation coefficients between the candidate data features are calculated;

[0020] Whether the candidate data features are strongly correlated is determined based on the Pearson correlation coefficients;

[0021] If the candidate data features are strongly correlated, the candidate data features are determined as redundant features;

[0022] The target data features with low importance in the redundant features are removed from the initial coupling model to obtain the reduced-order model.

[0023] In an implementation, the random forest includes a plurality of decision trees, and the importance of all candidate data features is calculated by the random forest, and includes:

[0024] The target contribution degree of each candidate data feature in each decision tree is determined.

[0025] According to the comprehensive calculation of the respective target contribution degrees, the total contribution degree of each candidate data feature is determined, and the importance degree includes the total contribution degree.

[0026] In an implementable manner, the determination of the contribution degree of each candidate data feature in each decision tree includes:

[0027] calculating a first Gini index of the candidate data feature at the first node of the i-th decision tree, a second Gini index of the candidate data feature at the second node of the i-th decision tree, and a third Gini index of the candidate data feature at the third node of the i-th decision tree, wherein the second node and the third node are branch nodes corresponding to the first node after the first node is split; k m l r l r m

[0028] determining the contribution degree of the candidate data feature at the first node of the i-th decision tree by using the first Gini index, the second Gini index, and the third Gini index; k m

[0029] comprehensively calculating the contribution degree of the candidate data feature at each node of the i-th decision tree, and determining a target contribution degree of the candidate data feature in the i-th decision tree. i k

[0030] In an implementable manner, the calculation of the Pearson correlation coefficients between the candidate data features includes:

[0031] determining the standard deviation of the candidate data features and the covariance between the candidate data features;

[0032] obtaining the Pearson correlation coefficients between the candidate data features by using a preset Pearson correlation coefficient algorithm, the standard deviation, and the covariance.

[0033] In an implementable manner, the total contribution degree is determined by using the following mathematical expression:

[0034]

[0035] In the formula, the total contribution degree of the candidate data feature VIM X i X i VIM k X i ​​​​​​​​​​​​​​​​is a candidate data feature X i In the first k contribution degree of a node of the i-th decision tree, K is the total number of decision trees, wherein,

[0036] ;

[0037] In the formula, VIM k,m ( X i is a candidate data feature X i In the first k contribution degree of a node of the i-th decision tree, m is the total number of decision trees, wherein, N In the number of splits on the decision tree k , wherein,

[0038] ;

[0039] In the formula, GI k,m ( X i is a first Gini index, GI k,l ( X i is a second Gini index, GI k,r ( X i is a third Gini index.

[0040] In one possible implementation, the Pearson correlation coefficient algorithm includes the following mathematical expression:

[0041] ;

[0042] In the formula, r ( X , Y ) is a Pearson correlation coefficient between a candidate data feature X and a candidate data feature Y , Cov ( X , Y ) represents a covariance between a candidate data feature X and a candidate data feature Y , Var [ X ] and Var [ Y ] respectively represent a candidate data feature X and a candidate data feature Y ​standard deviation.

[0043] To achieve the above object, the second aspect of the present application provides a transformer multi-physical field direct coupling reduced-order processing device, the device comprising:

[0044] A first modeling module is configured to obtain and utilize geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer.

[0045] A second modeling module is configured to construct an initial coupling model of the transformer based on heat generation, fluid flow and fluid-solid heat transfer process during operation of the transformer.

[0046] A reduced-order processing module is configured to utilize a random forest and the initial coupling model to establish a reduced-order model of the thermal-fluid-solid multi-physical field.

[0047] An operation simulation module is configured to perform multi-physical field operation simulation of the transformer according to the simulation geometric model and the reduced-order model to obtain an operation simulation result, which reflects the operating conditions of the transformer in the multi-physical field.

[0048] To achieve the above object, the third aspect of the present application provides a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the first aspect and any possible implementation.

[0049] To achieve the above object, the fourth aspect of the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the first aspect and any possible implementation.

[0050] The embodiments of the present application have the following beneficial effects:

[0051] The application provides a transformer multi-physical field direct coupling reduced-order processing method, which comprises the following steps: obtaining and utilizing the geometric parameters, material properties and operating conditions of a transformer to establish a simulation geometric model of the transformer; based on the heat generation, fluid flow and fluid-solid heat transfer process of the transformer during operation, an initial coupling model of the thermal-fluid-solid multi-physical field direct coupling of the transformer is constructed; a reduced-order model of the thermal-fluid-solid multi-physical field is established by using a random forest algorithm and the initial coupling model, and a multi-physical field operation simulation of the transformer is performed according to the simulation geometric model and the reduced-order model to obtain an operation simulation result, which is used to reflect the operating conditions of the transformer in the multi-physical field. By using the random forest algorithm and the initial coupling model, the reduced-order model of the thermal-fluid-solid multi-physical field is established, the initial coupling model is reduced, the complexity of the model is simplified, the degrees of freedom in the simulation model are reduced, the calculation efficiency of the simulation is greatly improved during the multi-physical field operation simulation, the demand for computing resources is reduced, and the reliability of the operation simulation result is maintained. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0053] Among them:

[0054] Figure 1 The flow chart of the transformer multi-physical field direct coupling reduced-order processing method in the embodiment of the present application is shown in FIG. 1.

[0055] Figure 2 Another flow chart of the transformer multi-physical field direct coupling reduced-order processing method in the embodiment of the present application is shown in FIG. 2.

[0056] Figure 3 The structural block diagram of the transformer multi-physical field direct coupling reduced-order processing device in the embodiment of the present application is shown in FIG. 3.

[0057] Figure 4 The structural block diagram of the computer device in the embodiment of the present application is shown in FIG. 4. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0059] To solve the problem that it is difficult to balance between calculation efficiency and accuracy when dealing with problems of mutual coupling of processes such as heat conduction, fluid flow and solid deformation, a reduced-order calculation method can be used. The reduced-order method simplifies the complexity of the model, reduces the number of variables required for calculation, and at the same time tries to maintain the key characteristics of the original physical process. Random forest is a classic ensemble learning method that can achieve classification and regression functions by training multiple decision trees, and can calculate the importance of each feature in the data set during training. By extracting the main variables that affect the behavior of the system according to certain rules, the degrees of freedom in the simulation model are reduced, the calculation efficiency of the simulation is greatly improved, the demand for computing resources is reduced, and the reliability of the results is maintained.

[0060] By effectively applying the reduced-order calculation method of multi-physical field direct coupling, the performance of the transformer under actual working conditions can be more accurately and efficiently predicted, which has important scientific and engineering application value for the design optimization, fault analysis, life prediction, digital twin model construction of the transformer, and the like. For details, please refer to the following contents.

[0061] Please refer to Figure 1 , Figure 1 The flowchart of a transformer multi-physical field direct coupling reduced-order processing method in the embodiment of the present application, which can be applied to a terminal and a server. The terminal can be a desktop terminal or a mobile terminal, and the mobile terminal can be at least one of a mobile phone, a tablet computer, a notebook computer, etc. The server can be implemented by an independent server or a server cluster composed of multiple servers. The embodiment is illustrated by taking application to a terminal as an example, and the method shown in Figure 1 includes the following steps:

[0062] 101. Obtain and use the geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer;

[0063] 102. Based on the heat generation, fluid flow and fluid-solid heat transfer processes during the operation of the transformer, an initial coupling model of the thermal-fluid-solid multi-physical field direct coupling of the transformer is constructed;

[0064] First, the physical characteristics and operating environment of the transformer need to be accurately captured in order to accurately restore the real environment and perform modeling, including but not limited to the geometric parameters, material properties and operating conditions of the transformer, etc. For example, the geometric dimensions of the transformer components, material properties (such as density, thermal conductivity, specific heat capacity, etc.) and working conditions (such as load current, ambient temperature and other environmental factors) are recorded in detail. These pre-determined actual data of the transformer are used as the basis for establishing a multi-physical field model to simulate the thermal behavior, fluid dynamics and structural response of the transformer in actual operation.

[0065] Firstly, the simulation geometric model of the transformer is established by using the geometric parameters, material properties and operating conditions of the transformer. Secondly, considering the processes of heat generation, fluid flow and fluid-solid heat transfer, the initial coupling model of the thermal-fluid-solid multi-physical field direct coupling of the transformer is constructed based on the heat generation, fluid flow and fluid-solid heat transfer processes of the transformer during operation, which is used as the thermal-fluid-solid multi-physical field coupling model of the transformer.

[0066] For example, the thermal-fluid-solid multi-physical field coupling model includes:

[0067] Considering the viscosity and pressure of the fluid, the fluid motion is described by the Navier-Stokes equation (1):

[0068] (1)

[0069] wherein, p is the density, v is the velocity field, p is the pressure, m is the dynamic viscosity, f is the volume force.

[0070] Considering the conduction, convection and radiation, the energy transfer in the fluid and solid is described by equation (2):

[0071] (2)

[0072] wherein, T is the temperature, c p is the specific heat capacity, k is the thermal conductivity, Q is the heat source term. The heat source term Q is derived from the internal heat source generated by chemical reaction or interaction with radiation, and the radiation heat transfer is described by the Stefan-Boltzmann law, as shown in equation (3):

[0073] (3)

[0074] wherein, σ is the Stefan-Boltzmann constant.

[0075] Further, the winding heat generation is described by Joule's law, as shown in equation (4):

[0076] (4)

[0077] Meanwhile, the propagation of heat will cause the solid components to expand thermally, thereby affecting the mechanical structure and electrical performance of the transformer. Considering the influence of fluid-structure heat transfer on the solid, and taking into account the effects of thermal stress and thermal expansion, a thermo-structure coupling equation is constructed, as shown in equation (5).

[0078] (5)

[0079] in, a It is the coefficient of thermal expansion. T 0 is the reference temperature. e T It is the thermal strain tensor, and u is the displacement field. s s It is the stress tensor. C is the elastic tensor, e is the total strain tensor, and I is the unit tensor.

[0080] Considering the normal stress (generated by fluid pressure) and shear stress (represented by the stress tensor) exerted by the fluid on the solid surface, the dot product of this combined tensor and the normal vector of the solid surface is used to obtain the total force acting on the solid surface, thereby establishing the fluid-structure interaction equation, as shown in equation (6):

[0081] (6)

[0082] in, F A It is the force exerted by a fluid on a solid surface. p It is fluid pressure. K It is the stress tensor, representing the shear stress and normal stress within the fluid. n It is the unit normal vector of the solid surface.

[0083] 103. Using random forest and the initial coupling model, establish a reduced-order model of the thermal-fluid-solid multiphysics field;

[0084] Furthermore, the order reduction model is used to perform order reduction calculations during the direct coupling of multiphysics calculations. Random forests can be used for order reduction; for example, the importance of all candidate data features can be calculated using a random forest, and order reduction can be performed based on the importance of these features, retaining those with high importance. The candidate data features are derived from pre-collected transformer measurement data. These candidate data features indicate the data features that affect the simulation results during the multiphysics simulation of the transformer, and the random forest is trained using the aforementioned measurement data as training sample data.

[0085] 104. Perform multi-physics field operation simulation of the transformer based on the simulation geometric model and the reduced-order model to obtain operation simulation results. The operation simulation results are used to reflect the operating conditions of the transformer in the multi-physics field.

[0086] Further, simulation can be performed according to the reduced-order model and the simulation geometric model, and multi-physical field operation simulation of the transformer is performed according to the simulation geometric model and the reduced-order model, so that operation simulation results reflecting the operation condition of the transformer in the multi-physical field are quickly obtained.

[0087] The application provides a reduced-order processing method for direct coupling of a transformer multi-physical field, which comprises the following steps: obtaining and using geometric parameters, material properties and an operation condition of a transformer to establish a simulation geometric model of the transformer; constructing an initial coupling model of thermal-hydraulic-structure direct coupling of the transformer multi-physical field based on heat generation, fluid flow and fluid-structure heat transfer processes during operation of the transformer; using a random forest algorithm and the initial coupling model to establish a reduced-order model of the thermal-hydraulic-structure multi-physical field; performing multi-physical field operation simulation of the transformer according to the simulation geometric model and the reduced-order model to obtain operation simulation results, which are used to reflect the operation condition of the transformer in the multi-physical field. By using the random forest algorithm and the initial coupling model to establish the reduced-order model of the thermal-hydraulic-structure multi-physical field, the initial coupling model is reduced, the complexity of the model is simplified, the degrees of freedom in the simulation model are reduced, the calculation efficiency of the simulation is greatly improved during the multi-physical field operation simulation, the demand for computing resources is reduced, and the reliability of the operation simulation results is maintained.

[0088] Please refer to Figure 2 , Figure 2 Another flowchart of the reduced-order processing method for direct coupling of a transformer multi-physical field in the embodiment of the application is shown in Figure 2 The method comprises the following steps:

[0089] 201. Obtain and use geometric parameters, material properties and an operation condition of a transformer to establish a simulation geometric model of the transformer.

[0090] 202. Construct an initial coupling model of thermal-hydraulic-structure direct coupling of the transformer multi-physical field based on heat generation, fluid flow and fluid-structure heat transfer processes during operation of the transformer.

[0091] It should be noted that steps 201 and 202 are similar to the contents of steps 101 and 102 shown in Figure 1 To avoid repetition, the contents of steps 101 and 102 shown in Figure 1 are not described here, and can be referred to

[0092] In a feasible implementation manner, in order to improve the calculation efficiency while ensuring the data accuracy, the random forest and the initial coupling model are utilized to establish the reduced order model of the thermal fluid-solid multi-physical field, and the two-stage feature reduction can be combined to determine whether different features are strongly correlated by combining the importance and correlation of the sample data features, if the correlation is strong, it is determined that the feature is redundant, and the feature with low importance is removed, and the two-stage feature reduction is combined to construct the reduced order model of the thermal fluid-solid multi-physical coupling field, so that the accuracy of the simulation can be ensured while reducing the feature dimension, wherein the importance of the data feature can be determined by the random forest, and the correlation can be determined by the Pearson correlation coefficient, for details, see below.

[0093] 203、calculating the importance of all candidate data features by the random forest, the candidate data features being used to indicate the data features affecting the operation simulation results in the operation simulation of the multi-physical field of the transformer;

[0094] It can be understood that the random forest is a classic ensemble learning method, which achieves the functions of classification and regression by training multiple decision trees, and can calculate the importance of each feature in the data set in the training process, the data set can be obtained from the measurement data, and the measurement data is used as sample data to obtain the subsequent training set, so the importance of all candidate data features is calculated by the random forest, and can be sorted from high to low. The implementation steps are as follows:

[0095] 1) Each time N training samples are extracted from the training set with replacement to form a new training set;

[0096] 2) A total of M times are extracted, and M decision trees are trained by using the new training set. In each node generated, d features are randomly selected without repetition, and the d features are used to divide the sample set;

[0097] 3) When performing a classification task, each decision tree outputs a classification result, and the final classification result is determined by voting of all decision trees.

[0098] The contribution of each data feature is evaluated by Gini Index. Gini Index represents the probability of a randomly selected sample being misclassified in the sample set. The Gini Index of the ith feature The Gini Index of the kth decision tree at node m As (7):

[0099] (7)

[0100] In the formula, C is the number of categories of classification, The proportion of the c-class at the node m, i.e. the probability of correct classification. The smaller the Gini index, the smaller the probability of misclassification of the selected sample in the set, i.e. the higher the purity of the set, and vice versa, the set is less pure.

[0101] Data feature The importance at (k, m) is represented by the change in the Gini index before and after the branch of node m VIM k,m （ ） , and the calculation method is as shown in equation (8):

[0102] (8)

[0103] In the formula: and represent the Gini indexes of the two new nodes after branching. l and r .

[0104] Data feature Split N times on the decision tree k, and sum all the data features that appear on the tree, then the contribution degree on the tree is as shown in equation (9):

[0105] (9)

[0106] If there are K trees in the random forest, then the total contribution degree in the random forest is as shown in equation (10):

[0107] (10)

[0108] Process each feature in the sample data, calculate the contribution degree of all features, and finally normalize it, so as to obtain the contribution degree score of each feature VIM ( X i ).

[0109] In a feasible implementation manner, the random forest includes a plurality of decision trees, and the step 203 includes steps A01 and A02:

[0110] A01, determining the target contribution degree of each candidate data feature in each decision tree;

[0111] In a feasible implementation manner, the step A01 can include steps B01, B02 and B03:

[0112] B01, Calculate the candidate data features in the first... k The first decision tree m The first Gini index of the node, the first l The second Gini index at the node and the first r The third Gini index of the nth node, the nth l The node and the first r The node is the _th m The branch nodes corresponding to each node after splitting;

[0113] B02. Using the first Gini index, the second Gini index, and the third Gini index, determine the candidate data features in the [missing information]. k The first decision tree m The contribution of each node;

[0114] B03, based on the candidate data features in the first... i The contribution of each node in the decision tree is comprehensively calculated to determine the candidate data feature in the th decision tree. k The target contribution of each decision tree.

[0115] A02. Based on the contribution of each target, a comprehensive calculation is performed to determine the total contribution of each candidate data feature, wherein the importance includes the total contribution.

[0116] In one feasible implementation, the total contribution is determined using the following mathematical expression:

[0117] ;

[0118] In the formula, VIM ( X i ) are candidate data features X i Total contribution VIM k ( X i ) are candidate data features X i In the k The target contribution of each decision tree. K Let be the total number of decision trees, where

[0119] ;

[0120] In the formula, VIM k,m ( X i ) are candidate data features X i In the k Nodes of a decision tree mcontribution degree of each data feature, N for the number of splits on the decision tree, k wherein,

[0121] ;

[0122] wherein, GI k,m X i is a first Gini index, GI k,l X i is a second Gini index, GI k,r X i is a third Gini index.

[0123] 204, calculating the Pearson correlation coefficient between each of the candidate data features;

[0124] Further, the Pearson correlation coefficient between each of the measurement data is calculated to form a Pearson correlation coefficient matrix, so as to obtain the Pearson correlation coefficient between each of the candidate data features. The Pearson correlation coefficient reflects the correlation between variables X and Y, and the value is between -1 and 1. The closer to 1, the stronger the positive correlation between the two variables, and the closer to -1, the stronger the negative correlation. Usually, the absolute value is taken to measure whether there is a correlation between two variables. Assuming that X and Y are two different nodes in a power distribution network, the power flow data of which is collected to obtain the Pearson correlation coefficient r of the two variables as formula (11):

[0125] (11)

[0126] In the formula, Cov(X, Y) represents the covariance of X and Y, Var[X] and Var[Y] represent the standard deviation of variables X and Y, respectively. The correlation between each pair of variables is measured by normalizing the result by dividing the covariance by the standard deviation.

[0127] That is, the Pearson correlation coefficient between each of the candidate data features is calculated, including determining the standard deviation of the candidate data features and the covariance between each of the candidate data features; and using a preset Pearson correlation coefficient algorithm, standard deviation and covariance to obtain the Pearson correlation coefficient between each of the candidate data features.

[0128] It can be understood that the Pearson correlation coefficient algorithm includes the following mathematical expression:

[0129] ;

[0130] In the formula,​​​r X , Y is a Pearson correlation coefficient between candidate data features X and candidate data features Y , Cov X , Y is a covariance between candidate data features X and candidate data features Y , Var [ X ] and Var [ Y ] respectively represent a standard deviation of candidate data features X and candidate data features Y .

[0131] 205. determining whether the candidate data features are strongly correlated with each other based on the Pearson correlation coefficient;

[0132] 206. if the candidate data features are strongly correlated with each other, determining that the candidate data features are redundant features;

[0133] 207. removing a target data feature with low importance from the initial coupling model among the redundant features to obtain a reduced model;

[0134] It should be noted that whether different measurements are strongly correlated is determined, and if the correlation is strong, the feature is determined to be a redundant feature, and the feature with low importance is removed. Specifically, first, the correlation coefficient is used to determine whether the features are strongly correlated. If there is a strong correlation, it means that there is redundancy, and the strongly correlated features can be removed. Then, the importance of the features is further evaluated, and the features with low importance are removed from the strongly correlated features (redundant features), so that the data accuracy is ensured and the order is reduced.

[0135] Through the above steps, the main features can be effectively extracted from the multivariate data, the dimension of the data set is reduced, and as much information as possible of the original data set is retained. Thus, the reduced order calculation equation under the transformer thermal-fluid-solid coupling is obtained.

[0136] Through the above steps, the transformer geometric model is established, the material parameters are imported, and the thermal-fluid-solid multi-physics field coupling equation is constructed. Then, the reduced order equation is constructed through standardization, eigenvalue calculation, eigenvector calculation, etc., and the thermal-fluid-solid multi-physics field fast calculation is performed.

[0137] 208. performing a multi-physics field operation simulation of the transformer according to the simulation geometric model and the reduced order model to obtain an operation simulation result, wherein the operation simulation result is used to reflect the operation condition of the transformer under the multi-physics field. ​​

[0138] It should be noted that step 208 and Figure 1 The content of step 104 shown is similar, and will not be repeated here to avoid repetition. For details, please refer to [link / reference needed]. Figure 1 The content of step 104 shown.

[0139] This invention provides a method for reducing the order of direct coupling of multiphysics fields in transformers, which can be viewed as a method for reducing the order of direct coupling of thermal-fluid-structure interaction (TFI) multiphysics fields in transformers. The method involves acquiring the physical characteristics and actual operating conditions of the transformer, and establishing a corresponding simulation geometric model in simulation software, considering the transformer's geometric dimensions and material parameters. A TFI multiphysics model is established, taking into account heating, fluid flow, and fluid-structure heat transfer processes. A two-stage feature reduction method is used to combine the importance and correlation of sample data features, removing features with insufficient contribution, and constructing a reduced-order model of the TFI multiphysics coupling field. The geometric and physical models are then combined to perform TFI multiphysics direct coupling order reduction calculations. This invention comprehensively considers physical processes such as heating, fluid flow, and fluid-structure heat transfer to establish a direct coupling model of TFI multiphysics fields in transformers. By establishing a TFI multiphysics order reduction model and utilizing a two-stage feature reduction method, the computational dimensions are reduced, enabling rapid calculation of multiphysics fields under direct TFI direct coupling.

[0140] Please see Figure 3 , Figure 3 This is a structural block diagram of a transformer multi-physics direct coupling order reduction processing device according to an embodiment of the present invention, as shown below. Figure 3 The apparatus shown includes:

[0141] First modeling module 301: used to acquire and utilize the geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer;

[0142] The second modeling module 302 is used to construct an initial coupling model of the direct coupling of thermal, fluid flow and fluid-structure heat transfer processes of the transformer during operation.

[0143] The order reduction processing module 303 is used to establish a reduced-order model of the thermal-fluid-solid multiphysics field using random forest and the initial coupling model.

[0144] The simulation module 304 is used to perform multi-physics field operation simulation of the transformer based on the simulation geometric model and the reduced-order model, and obtain the operation simulation results. The operation simulation results are used to reflect the operating conditions of the transformer in the multi-physics field.

[0145] It should be noted that, as Figure 3 The device shown is Figure 1The content of each step in the method is similar, and to avoid repetition, it is not described here, and the specific content can be referred to Figure 1 The content of each step in the method.

[0146] The application provides a transformer multi-physical field direct coupling reduction processing device, which comprises: a first modeling module: used for acquiring and utilizing the geometric parameters, material properties and operating conditions of a transformer, and establishing a simulation geometric model of the transformer; a second modeling module: used for constructing an initial coupling model of the thermal-hydraulic-mechanical multi-physical field direct coupling of the transformer based on the heat generation, fluid flow and fluid-solid heat transfer process of the transformer during operation; a reduction processing module: used for establishing a reduction model of the thermal-hydraulic-mechanical multi-physical field by using a random forest and the initial coupling model; and a running simulation module: used for performing multi-physical field running simulation of the transformer according to the simulation geometric model and the reduction model, and obtaining a running simulation result, wherein the running simulation result is used to reflect the operating conditions of the transformer in the multi-physical field. By using the random forest algorithm and the initial coupling model, the reduction model of the thermal-hydraulic-mechanical multi-physical field is established, the reduction of the initial coupling model is realized, the complexity of the model is simplified, thereby reducing the degrees of freedom in the simulation model, the calculation efficiency of the simulation can be greatly improved during the multi-physical field running simulation, the demand for computing resources is reduced, and the reliability of the running simulation result is maintained.

[0147] Figure 4 An internal structure diagram of a computer device in an embodiment is shown. The computer device can be a terminal or a server. As shown in the figure, Figure 4 The computer device comprises a processor, a memory and a network interface connected through a system bus. The memory comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system, and can also store a computer program, which, when executed by the processor, can enable the processor to implement the above method. The internal memory can also store a computer program, which, when executed by the processor, can enable the processor to execute the above method. Those skilled in the art can understand that Figure 4 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can comprise more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0148] In one embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to enable the processor to perform the steps as shown in Figure 1 Or Figure 2 The steps shown.

[0149] In one embodiment, a computer readable storage medium is provided, storing a computer program, which, when executed by a processor, causes the processor to perform the steps shown in Figure 1 or Figure 2

[0150] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a non-volatile computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0151] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0152] The above embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.​

Claims

1. A method for reducing the order of direct coupling of multi-physics fields in a transformer, characterized in that, The method comprises: acquiring and utilizing geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer, wherein the material properties at least include density, thermal conductivity and specific heat capacity, and the operating conditions at least include load current and ambient temperature; constructing an initial coupling model of thermal-hydraulic-solid multi-physical field direct coupling of the transformer based on heat generation, fluid flow and fluid-solid heat transfer process during operation of the transformer; establishing a reduced order model of the thermal-hydraulic-solid multi-physical field by utilizing random forest and the initial coupling model; performing multi-physical field operation simulation of the transformer according to the simulation geometric model and the reduced order model to obtain operation simulation results, which are used to reflect operation conditions of the transformer in the multi-physical field; wherein the step of establishing the reduced order model of the thermal-hydraulic-solid multi-physical field by utilizing random forest and the initial coupling model comprises: calculating the importance of all candidate data features by random forest, wherein the candidate data features are used to indicate data features affecting operation simulation results during the multi-physical field operation simulation of the transformer; calculating Pearson correlation coefficients between each two of the candidate data features; judging whether each two of the candidate data features are strongly correlated based on the Pearson correlation coefficients; if each two of the candidate data features are strongly correlated, determining that each two of the candidate data features are redundant features; eliminating target data features with low importance from the initial coupling model to obtain a reduced order model; wherein the random forest comprises a plurality of decision trees, and the step of calculating the importance of all candidate data features by random forest comprises: determining target contribution degrees of each candidate data feature in each decision tree; comprehensively calculating the target contribution degrees to determine total contribution degrees of each candidate data feature, wherein the importance comprises the total contribution degrees.

2. The method of claim 1, wherein, the step of determining the target contribution degrees of each candidate data feature in each decision tree comprises: Calculate the candidate data features in the first... k The first decision tree m The first Gini index of the node, the first l The second Gini index at the node and the first r The third Gini index of the nth node, the nth l The node and the first r The node is the _th m The branch nodes corresponding to each node after splitting; The first Gini index, the second Gini index, and the third Gini index are used to determine a contribution degree of the candidate data feature at a first node of a decision tree. k The first Gini index, the second Gini index, and the third Gini index are used to determine a contribution degree of the candidate data feature at a first node of a decision tree. m The first Gini index, the second Gini index, and the third Gini index are used to determine a contribution The contribution degree of each node of the decision tree is calculated, and the target contribution degree of the candidate data feature in the decision tree is determined. i The contribution degree of each node of the decision tree is calculated, and the target contribution degree of the candidate data feature in the decision tree is determined. k The contribution degree of each node of the decision tree is calculated, and the target contribution degree of the candidate data feature in the decision tree is determined.

3. The method of claim 1, wherein, the step of calculating the Pearson correlation coefficients between each two of the candidate data features comprises: determining standard deviations of the candidate data features and covariances between each two of the candidate data features; obtaining the Pearson correlation coefficients between each two of the candidate data features by utilizing a preset Pearson correlation coefficient algorithm, the standard deviations and the covariances.

4. The method of claim 2, wherein, The total contribution degree is determined by the following mathematical expression: ； wherein, VIM X i is the total contribution of the candidate data feature X i , VIM k X i is the target contribution of the candidate data feature X i the first decision tree, k K is the total number of decision trees, wherein,​​​ ； wherein VIM k,m ( X i ) is a candidate data feature X i In the first k contribution of a node of a decision tree, m N is number of splits on a decision tree k , wherein,​ ； wherein GI k,m X i is a first Gini index, GI k,l X i is a second Gini index, GI k,r X i is a third Gini index.​​​ 5. The method of claim 3, wherein, The Pearson correlation coefficient algorithm comprises the following mathematical expressions: ； In the formula: r ( X , Y ) are candidate data features X and candidate data features Y The Pearson correlation coefficient between them Cov ( X , Y ) represents the candidate data features X and candidate data features Y Covariance between Var [ X ]and Var [ Y ] respectively represent the features of the candidate data X and candidate data features Y The standard deviation.

6. A reduced-order processing device for direct coupling of multi-physics in a transformer, the device comprising: a transformer model; a plurality of physics models; and a reduced-order model (ROM) coupling the transformer model and the plurality of physics models. The device comprises: a first modeling module configured to acquire and utilize geometric parameters, material properties and operating conditions of the transformer to establish a simulation geometric model of the transformer; a second modeling module configured to construct an initial coupling model of thermal-hydraulic-solid multi-physical field direct coupling of the transformer based on heat generation, fluid flow and fluid-solid heat transfer process during operation of the transformer; a reduced order processing module configured to establish a reduced order model of the thermal-hydraulic-solid multi-physical field by utilizing random forest and the initial coupling model; The operation simulation module is configured to perform multi-physical field operation simulation of the transformer according to the simulation geometric model and the reduced-order model, and obtain an operation simulation result, which reflects an operation condition of the transformer in the multi-physical field. The reduced-order processing module is specifically configured to: calculate the importance of all candidate data features by using the random forest, the candidate data features being used to indicate data features that affect the operation simulation result in the multi-physical field operation simulation of the transformer; calculate the Pearson correlation coefficient between each two of the candidate data features; determine whether each two of the candidate data features are strongly correlated based on the Pearson correlation coefficient; if each two of the candidate data features are strongly correlated, determine that each two of the candidate data features are redundant features; and remove a target data feature with low importance from the redundant features from the initial coupled model to obtain the reduced-order model. The random forest includes a plurality of decision trees, and the calculation of the importance of all candidate data features by using the random forest includes: determining a target contribution degree of each candidate data feature in each decision tree; and determining a total contribution degree of each candidate data feature according to the target contribution degrees, the importance including the total contribution degree.

7. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to enable the processor to perform the steps of the method according to any one of claims 1 to 5. 8.A computer device, comprising a memory and a processor, and characterized in that, The memory stores a computer program, and the computer program is executed by the processor to enable the processor to perform the steps of the method according to any one of claims 1 to 5.

Images