Pouring optimization method and system for transformer winding

Abstract

The invention relates to the technical field of transformers, and provides a transformer winding pouring optimization method and system, and the method comprises the steps: constructing a three-dimensional model of a transformer winding, and obtaining a pouring parameter space; randomly generating a first pouring parameter in the parameter space, and performing bubble prediction and insulating layer thickness distribution in combination with the model to obtain a bubble parameter and thickness distribution; carrying out distortion identification on the distribution of the insulating layer to obtain a distortion degree, and calculating the pouring fitness by combining bubble parameters; and according to the bubble parameter and the distortion degree, configuring a pouring adjustment step length, adjusting the first pouring parameter, continuously optimizing the parameters, finally obtaining an optimal pouring parameter, and carrying out winding pouring. The technical problems that bubbles are generated and the thickness of an insulating layer is uneven in the transformer winding pouring process are solved, and the technical effects that by dynamically adjusting parameters in the pouring process, generation of the bubbles is avoided, and the pouring quality and performance of the transformer winding are improved are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of transformers, and in particular to a method and system for optimizing the casting of transformer windings. Background Art

[0002] Transformers are important equipment in power systems, and the quality of their windings directly affects the performance and reliability of transformers. In the traditional transformer winding manufacturing process, the casting process of the winding has a vital impact on the insulation performance and mechanical strength of the transformer. At present, the casting technology of transformer windings still faces many challenges, especially in terms of bubble control and uniformity of insulation layer thickness. The presence of bubbles may lead to uneven insulation layer, which in turn affects the thermal stability and electrical performance of the transformer; while the uneven thickness of the insulation layer may lead to a decrease in the mechanical strength of the winding, and even failure under high-voltage working conditions. Traditional casting processes rely more on manual experience and the setting of quantitative parameters, but lack effective means for real-time monitoring and adjustment of the casting process. Traditional casting methods cannot be optimized and adjusted for each batch of windings, and it is difficult to achieve more efficient and accurate production requirements. In addition, with the increasing complexity of the transformer application environment, the winding is required to maintain an efficient and stable working state in a wider temperature and voltage range. Traditional processes and equipment have been difficult to meet these requirements. Summary of the invention

[0003] The present application provides a method and system for optimizing the casting of transformer windings, aiming to solve the technical problems of bubble generation and uneven thickness of the insulation layer during the casting process of the transformer windings.

[0004] In view of the above problems, the present application provides a method and system for optimizing the casting of transformer windings.

[0005] The first aspect disclosed in the present application provides a method for optimizing the casting of a transformer winding, the method comprising: constructing a three-dimensional model of a transformer winding to be cast, and obtaining a casting parameter space; randomly generating a first casting parameter in the casting parameter space, and combining the three-dimensional model of the winding to predict bubbles after casting and generate an insulation layer thickness distribution, to obtain a first bubble parameter and a first insulation layer thickness distribution; performing a distortion recognition analysis on the first insulation layer thickness distribution to obtain a first distortion degree, and calculating a first casting fitness in combination with the first bubble parameter; configuring a casting adjustment step according to the first bubble parameter and the first distortion degree, adjusting the first casting parameter, continuing to optimize the casting parameter, obtaining an optimal casting parameter, and casting the transformer winding.

[0006] Another aspect disclosed in the present application provides a casting optimization system for a transformer winding, the system comprising: constructing a three-dimensional model of a transformer winding to be cast, and obtaining a casting parameter space; randomly generating a first casting parameter in the casting parameter space, and combining the three-dimensional model of the winding to predict bubbles after casting and generate an insulation layer thickness distribution, to obtain a first bubble parameter and a first insulation layer thickness distribution; performing a distortion recognition analysis on the first insulation layer thickness distribution to obtain a first distortion degree, and calculating a first casting fitness in combination with the first bubble parameter; configuring a casting adjustment step according to the first bubble parameter and the first distortion degree, adjusting the first casting parameter, continuing to optimize the casting parameter, obtaining an optimal casting parameter, and casting the transformer winding.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The above-mentioned method for optimizing the casting of transformer windings first constructs a three-dimensional model of the transformer winding to obtain the relevant casting parameter space, which provides a basis for subsequent optimization; then, randomly generates preliminary casting parameters in the casting parameter space, and simulates in combination with the winding three-dimensional model to predict the bubble distribution and the change in the thickness of the insulation layer after casting, which provides preliminary data for subsequent optimization; then, the first distortion degree is obtained by using the distortion recognition analysis of the thickness distribution of the first insulation layer to identify the uneven areas or potential structural problems in the distribution of the insulation layer, and then the casting fitness calculated by combining the first bubble parameter helps to further evaluate the advantages and disadvantages of the casting scheme during the optimization process, and avoid insulation layer defects or structural instability caused by improper casting; finally, according to the comprehensive analysis of the first bubble parameter and the distortion degree, the step size of the casting parameters is adjusted, and the casting scheme is gradually optimized. Through repeated adjustment and optimization, the optimal casting parameters are finally obtained, thereby ensuring the quality of transformer winding casting. This optimization process can significantly reduce problems caused by inappropriate parameters, such as excessive bubbles or uneven insulation layer, and improve the stability and reliability of the casting process, thereby ensuring the long-term performance and safety of the transformer winding.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 The figure is a flow chart of a method for optimizing casting of transformer windings in one embodiment.

[0011] Figure 2 The present invention is a system architecture diagram for optimizing the casting of transformer windings in one embodiment.

[0012] Explanation of the reference numerals: parameter space acquisition module 1, parameter distribution acquisition module 2, parameter distribution acquisition module 3, transformer winding casting module 4. DETAILED DESCRIPTION

[0013] The embodiments of the present application provide a method and system for optimizing the casting of transformer windings, thereby solving the technical problems of bubble generation and uneven thickness of the insulation layer during the casting of transformer windings.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] Embodiment 1, as Figure 1 As shown, the present application provides a method for optimizing the casting of transformer windings, the method comprising: Construct a three-dimensional winding model of the transformer winding to be cast, and obtain the casting parameter space.

[0017] In an embodiment of the present application, the system terminal first needs to construct a three-dimensional model of the transformer winding to be cast. This process first collects the detailed geometric shape of the winding, including the arrangement and shape of the wires on the winding surface. Since the wires on the winding are usually uneven, the complexity of this structure requires that the casting process can accurately adapt to these changing surfaces; by establishing a three-dimensional model of the winding, the spatial distribution and morphological characteristics of each wire in the winding can be accurately described, providing accurate basic data for subsequent casting optimization; based on this three-dimensional model, the system terminal will obtain the casting parameter space. The casting parameter space is a space that contains a parameter about the casting speed. The casting speed directly affects the injection process and flow state of the resin. When the winding surface is irregular, the casting speed needs to be precisely controlled to ensure that the resin can be smoothly filled into all gaps to avoid the generation of bubbles or insufficient resin infiltration.

[0018] Furthermore, the present application provides a method for constructing a three-dimensional winding model of a transformer winding to be cast, and obtaining a casting parameter space, including: The surface morphology of the transformer winding to be cast is collected to construct a three-dimensional winding model of the transformer winding; the casting speed adjustment space for casting the transformer winding is obtained and constructed as a casting parameter space.

[0019] Preferably, the system terminal uses a high-precision 3D scanner or other surface acquisition equipment to scan the transformer winding to be cast. This process can obtain the geometric shape of the winding surface through laser scanning or optical measurement technology. The scanning equipment can capture the shape, size, and surface unevenness of the winding wire. The surface data obtained through scanning is converted into point cloud data. These point cloud data represent the spatial coordinate information of each point on the winding surface; then, these point cloud data are input into 3D modeling software (such as AutoCAD, SolidWorks, etc.). The 3D modeling software obtains a complete winding 3D model by fitting and processing the point cloud data. The model can accurately reflect the size, shape and surface roughness of the winding, and provide basic data for the simulation and optimization of the subsequent casting process; after the winding 3D model is completed, it is necessary to build a casting speed adjustment space for the casting process. This space involves a control tree The key parameter of resin flow is the pouring speed. The pouring speed affects the flow behavior of the resin, and then affects whether the resin can evenly penetrate into various parts of the winding. On the concave and convex surface of the winding wire, the speed of resin flow needs to be precisely controlled to avoid uneven pouring results. To this end, by analyzing the geometric shapes of different areas in the winding three-dimensional model and the physical properties of resin flow during the pouring process, a pouring speed adjustment space is constructed. This space contains the factors affecting the flow and distribution of resin at different pouring speeds. The construction of the pouring speed adjustment space needs to be combined with physical simulation software to simulate the flow of resin at different speeds and generate a parameter space that can be adjusted in real time for subsequent optimization control; finally, by using the constructed pouring speed adjustment space as the pouring parameter space, the construction of this pouring parameter space is the basis of the entire optimization, which lays a solid theoretical and data foundation for the automatic adjustment and optimization of parameters in the optimization process.

[0020] A first casting parameter is randomly generated in the casting parameter space, and combined with the winding three-dimensional model, bubble prediction and insulation layer thickness distribution after casting are performed to obtain a first bubble parameter and a first insulation layer thickness distribution.

[0021] In one embodiment, the system terminal randomly generates a pouring speed from the pouring parameter space as the first pouring parameter. The first pouring parameter is randomly generated to explore different possibilities in the pouring parameter space. During the pouring process, the resin will flow and cover the winding surface. The unevenness and irregularity of the winding surface (such as the unevenness of the wire) will affect the resin infiltration effect, resulting in differences in the thickness of the insulation layer at different locations. In addition, the change of pouring speed will also affect the fluidity of the resin, which may lead to incomplete resin filling in some areas, resulting in bubbles or uneven thickness of the insulation layer. In order to accurately analyze this process, the system terminal predicts the number of bubbles that may be generated during the pouring process based on the first pouring parameters and the winding three-dimensional model through the established bubble prediction path, and obtains the first bubble parameter. The generation of these bubbles is usually caused by poor resin flow or air entrainment. These bubbles will affect the density and uniformity of the insulation layer. At the same time, based on the first pouring parameters and the winding three-dimensional model, the thickness distribution of the insulation layer after pouring will also be generated. This step simulates the infiltration of the resin at different positions through the insulation layer thickness generation path, and generates a predicted insulation layer thickness distribution as the first insulation layer thickness distribution. This first insulation layer thickness distribution can accurately reflect the thickness difference of the insulation layer at different positions, especially in the uneven area of ​​the winding surface. Through these analysis processes, the first bubble parameter and the first insulation layer thickness distribution can be obtained. These data will help the system terminal understand the potential problems in the pouring process, and then provide a basis for optimizing the pouring parameters, and ultimately achieve the goal of improving the quality of the transformer winding.

[0022] Furthermore, the present application provides a method for randomly generating a first casting parameter in the casting parameter space, combining the winding three-dimensional model, performing bubble prediction after casting and generating an insulation layer thickness distribution, and obtaining a first bubble parameter and a first insulation layer thickness distribution, including: A first casting parameter is randomly generated in the casting parameter space; the first casting parameter and the winding three-dimensional model are input into a pre-constructed bubble prediction path to predict and obtain the first bubble parameter, wherein the bubble prediction path is constructed by training with sample casting parameters, a sample winding three-dimensional model set and a sample bubble parameter set, and each sample bubble parameter includes the number of sample bubbles; according to the winding three-dimensional model and the first casting parameter, the thickness distribution of the insulation layer after casting is generated to obtain the first insulation layer thickness distribution.

[0023] Preferably, the system terminal obtains the first casting parameter and the constructed winding three-dimensional model as input data, and inputs them into a pre-constructed bubble prediction path. The bubble prediction path is established by using machine learning, and is specifically used to predict the bubble parameters generated under different casting conditions. This path is obtained by combining a sample casting parameter set (including a sample casting speed), a sample winding three-dimensional model set (including different types of winding structures and surface morphologies of samples) and a sample bubble parameter set (including the number of bubbles generated under different sample casting conditions). Taking the 3D convolutional neural network (3D-CNN) in machine learning as an example, the system terminal constructs a bubble prediction path based on the 3D convolutional neural network, including an input layer, a 3D convolution layer, a pooling layer, a fully connected layer and an output layer, and then through voxelization (Voxelization) and point cloud interpolation (Point Cloud The sample winding 3D model set is converted into a sample winding voxel data set by methods such as CNN and CNN, and together with the sample casting parameter set and the sample bubble parameter set, it forms a training set, a validation set and a test set; then, a random initialization method is used to set the convolution kernel of the 3D convolution layer and the weight of the fully connected layer, and then the training set is input into the initialized bubble prediction path for forward propagation. The data is passed layer by layer through the input layer, 3D convolution layer, pooling layer, fully connected layer, etc. to generate predicted bubble parameters; then, the loss value between the predicted bubble parameters and the sample bubble parameters is quantified by the mean square error (MSE) loss function, and the gradient of the loss to the weight of each layer is calculated layer by layer through the back propagation algorithm. The path parameters are optimized using the Adam optimizer, and the weights are adjusted to minimize the value of the loss function, and to ensure that the bubble prediction path gradually converges. This process will be carried out for multiple training cycles (Epochs). After each cycle, the performance of the bubble prediction path is tested using the validation set, and the MSE on the validation set is calculated to evaluate the bubble prediction. The accuracy of the test path in the bubble number prediction task is measured. If the loss on the validation set is not significantly reduced, the bubble prediction path can be further optimized by adjusting hyperparameters such as the learning rate and batch size. After the verification is passed, the current bubble prediction path is tested using the test set. If the bubble prediction path can achieve the expected accuracy on the test set, the training process is completed and the current bubble prediction path is output. Otherwise, the path structure is further adjusted or data enhancement technology is used to improve the training effect. During use, this bubble prediction path can predict the number of bubbles generated in the actual pouring process based on the input winding three-dimensional model and the first pouring parameter, thereby obtaining the first bubble parameter. In addition, the system terminal analyzes the insulation layer thickness distribution after pouring the winding three-dimensional model and the first pouring parameter through the insulation layer thickness generation path constructed based on the generative adversarial network, and generates the first insulation layer thickness distribution based on the analysis results to reflect the distribution of resin on the winding, thereby ensuring the uniformity and quality of the insulation layer.

[0024] Further, the present application provides generating a thickness distribution of the insulation layer after casting according to the winding three-dimensional model and the first casting parameter to obtain a first insulation layer thickness distribution, including: According to the transformer winding casting detection data in historical time, a sample casting parameter set and a sample winding three-dimensional model set are collected, and the insulation layer thicknesses at multiple positions of the transformer winding after casting with different sample casting parameters and sample winding three-dimensional models are collected to obtain a sample insulation layer thickness distribution set; based on a generative adversarial network, an insulation layer thickness generation path is constructed, wherein the insulation layer thickness generation path includes a generator and a discriminator; the sample casting parameter set, the sample winding three-dimensional model set and the sample insulation layer thickness distribution set are used as supervised training data to train the generator and the discriminator until convergence to obtain an insulation layer thickness generation path; the winding three-dimensional model and the first casting parameter are input into the insulation layer thickness generation path to generate a first insulation layer thickness distribution.

[0025] Optionally, based on the casting detection data of the transformer winding in the historical time, the system terminal collects and obtains the sample casting parameter set and the sample winding three-dimensional model set, and also collects the insulation layer thickness data of multiple positions of the transformer winding after casting with different sample casting parameters and winding three-dimensional models, to form a sample insulation layer thickness distribution set. These three sets will provide supervision data for the subsequent generative adversarial network (GAN) training; then, an insulation layer thickness generation path is constructed based on the generative adversarial network (GAN), and the generation path includes a generator (Generator) and a discriminator (Discriminator). The task of the generator is to generate a predicted insulation layer thickness distribution based on the input casting parameters and winding three-dimensional model. Its goal is to deceive the discriminator so that it cannot distinguish between the real and generated thickness data. The discriminator is used to determine whether the generated thickness distribution is real. Its goal is to distinguish between the generated false data and the real data, so that the thickness distribution generated by the generator is close enough to the actual situation. The generator and the discriminator are continuously optimized through adversarial training until the generator can generate an insulation layer thickness distribution that is highly consistent with the actual data; in the training During the process, the system terminal starts with the prepared sample data, and uses the sample casting parameter set, the sample winding three-dimensional model set and the sample insulation layer thickness distribution set as supervised training data. These data provide input information for the generator to help the generator learn the mapping relationship from casting conditions and winding models to insulation layer thickness distribution. At the beginning of the training process, the discriminator is trained first. In each training batch, the discriminator receives the real insulation layer thickness distribution data and the false data generated by the generator. The discriminator outputs a probability value by judging these two types of data, indicating the possibility that the data is real data. In this process, the discriminator uses the cross entropy loss function to measure the accuracy of its judgment, and calculates the gradient through back propagation to update the weight; then train the generator. The input of the generator is the winding three-dimensional model and casting parameters, and the output is the generated insulation layer thickness distribution. The generator does not directly calculate the error between the generated data and the real data, but adjusts the generated result through the feedback of the discriminator. Specifically, the generator updates the weight by maximizing the discriminator's authenticity score for its generated data, so as to make the generated thickness distribution more realistic.The loss function of the generator is calculated based on the output of the discriminator. The generator is continuously optimized, and eventually the generated thickness distribution becomes closer and closer to the actual data. This training process is carried out alternately. Each time training, the discriminator is optimized first, and the generator is trained through the feedback of the discriminator. The generator gradually adjusts the generation process through the judgment of the discriminator to produce more realistic results. In this adversarial process, the generator and the discriminator compete with each other. The discriminator tries to distinguish between true and false data, while the generator tries to make it impossible for the discriminator to distinguish between the false data and real data it generates. The adversarial optimization of the generator and the discriminator will continue until the thickness distribution generated by the generator is realistic enough and the discriminator finds it difficult to make an accurate judgment. The training process will last for multiple cycles (Epochs). At the end of each cycle, the generator and the discriminator are tested using the validation set. The performance of the discriminator is evaluated. Through the evaluation, it is determined whether the thickness distribution generated by the generator is gradually approaching the real data, ensuring the training effect of the generator and the discriminator. As the training progresses, the generator will be continuously optimized to generate more and more realistic thickness distributions, while the discriminator will gradually improve its ability to distinguish true and false data. When the generator and the discriminator reach an adversarial balance, the training process is completed. After the training, the generator can accurately predict the corresponding insulation layer thickness distribution based on the new casting parameters and the winding three-dimensional model. At this time, the system terminal inputs the winding three-dimensional model to be predicted and the first casting parameters into the trained generator to generate the first insulation layer thickness distribution. This prediction result provides a basis for subsequent casting optimization and helps adjust the key parameters in the casting process, thereby ensuring that the insulation layer thickness of the transformer winding is uniform and meets the quality requirements.

[0026] A distortion recognition analysis is performed on the thickness distribution of the first insulating layer to obtain a first distortion degree, and a first casting adaptability is calculated in combination with the first bubble parameter.

[0027] In one embodiment, after generating the first insulation layer thickness distribution, the next step is to perform distortion identification analysis on the distribution to ensure that the generated insulation layer thickness meets the expected quality standard. Due to the uneven thickness on the winding surface, the insulation layer may be too thin or too thick in some areas. Therefore, the system terminal needs to calculate the distortion degree to measure the unevenness of the insulation layer thickness. The calculation basis of the distortion degree is the thickness distribution of the insulation layer surface itself. The degree of unevenness is quantified by analyzing the depth difference between the recessed area and the convex area of ​​the transformer winding and the thickness difference of the first insulation layer thickness distribution, thereby obtaining the first distortion degree. This first distortion degree is used to describe the unevenness of the insulation layer. The larger the distortion degree, the more uneven the thickness of the insulation layer, which may affect the performance and safety of the transformer. Subsequently, the calculated first distortion degree is combined with the first bubble parameter to calculate the first casting fitness through the fitness calculation formula to reflect the quality of the current casting process. The higher the fitness, the better the quality of the insulation layer, and the smaller the impact of bubbles and thickness distortion on the quality. Otherwise, the parameters of the casting process need to be adjusted to improve the casting effect to ensure that the final transformer winding reaches the ideal quality standard.

[0028] Furthermore, the present application provides a method of performing a distortion identification analysis on the thickness distribution of the first insulating layer to obtain a first distortion degree, and calculating a first pouring adaptability in combination with the first bubble parameter, including: According to the winding three-dimensional model, the maximum depth difference between the recessed area and the raised area of ​​the transformer winding is extracted; according to the first insulating layer thickness distribution, the difference between the minimum insulating layer thickness and the maximum insulating layer thickness is calculated to obtain the first insulating layer thickness difference; the ratio of the first insulating layer thickness difference to the maximum depth difference is calculated as the first distortion degree; according to the first distortion degree and the first bubble parameter, the first casting adaptability of the first casting parameter is calculated.

[0029] Optionally, the system terminal extracts the characteristics of the winding surface morphology, especially the depression depth of the depression area and the protrusion height of the protrusion area, based on the winding three-dimensional model, and obtains the maximum depth difference of the transformer winding by adding the maximum protrusion height to the maximum depression depth. This maximum depth difference reflects the degree of unevenness of the winding surface geometry; then, based on the first insulation layer thickness distribution, the minimum insulation layer thickness and the maximum insulation layer thickness are extracted, and the maximum insulation layer thickness and the minimum insulation layer thickness are used to perform a difference calculation to obtain the first insulation layer thickness difference. The purpose of this step is to evaluate the thickness variation range of the insulation layer, that is, to identify the difference between the thinnest and thickest areas in the insulation layer. This difference indicates The uniformity of the insulation layer, the larger the difference, the higher the degree of non-uniformity of the insulation layer; then, the first insulation layer thickness difference is compared with the maximum depth difference, and their ratio is calculated as the first distortion. The larger the first distortion, the greater the unevenness on the winding surface and the greater the thickness difference of the insulation layer, indicating that the casting process may be relatively uneven. The lower the distortion, the smaller the unevenness of the winding surface, and the more uniform the thickness of the insulation layer; finally, the first distortion and the first bubble parameter are input into the fitness calculation formula to calculate the first casting fitness. The higher the fitness, the more uniform the casting process, the fewer bubbles, and the smaller the thickness difference of the insulation layer. On the contrary, it means that the casting quality is poor and the parameters need to be optimized. Through this series of calculation steps, the fitness of the casting process can be finally obtained, which provides a basis for adjusting and optimizing the casting parameters and ensuring the uniformity and quality of the insulation layer of the transformer winding.

[0030] Furthermore, the present application provides a method for calculating a first pouring fitness of a first pouring parameter according to the first distortion degree and the first bubble parameter, as shown in the following formula: ; Among them, JZF is the casting adaptability, is the bubble weight, is the distortion weight, and The sum of is 1, is the bubble parameter, is the bubble number threshold value of transformer winding casting, K is the thickness variance of the first insulation layer thickness distribution, and J is the distortion degree.

[0031] Optionally, the system terminal inputs the first distortion degree and the first bubble parameter into a fitness calculation formula. The specific content of the fitness calculation formula is as follows: ; Among them, JZF is the casting adaptability, which is used to comprehensively evaluate the effect of the casting process. is the bubble weight, which indicates the proportion of bubble parameters in fitness evaluation. is the distortion weight, which indicates the proportion of distortion in fitness evaluation. and The sum of is 1, is the bubble parameter, which indicates the number of bubbles in the actual pouring process. is the threshold value of the number of bubbles in the transformer winding casting (determined based on business needs), K is the thickness variance of the thickness distribution of the first insulation layer, reflecting the uniformity of the thickness distribution. The smaller the variance, the more uniform the distribution. J is the distortion, reflecting the relationship between the winding geometry and the uneven thickness of the insulation layer. Through this formula, the first casting adaptability comprehensively considers the two main factors of bubble generation and thickness distribution during the casting process, thereby providing a quantitative basis for optimizing the casting process parameters.

[0032] The casting adjustment step is configured according to the first bubble parameter and the first distortion degree, the first casting parameter is adjusted, the casting parameter is continuously optimized, the optimal casting parameter is obtained, and the casting of the transformer winding is performed.

[0033] In one embodiment, during the optimization of the casting parameters, a suitable casting adjustment step is configured according to the first bubble parameter and the first distortion degree so as to iteratively adjust the current casting parameters. The configuration of the adjustment step is intended to balance the optimization speed and accuracy, ensuring that the parameter adjustment can gradually approach the optimal solution, while avoiding oscillation caused by excessive step size or slow convergence caused by excessive step size. Specifically, the adjustment step size for each round of optimization is calculated according to the bubble parameters and the distortion degree of the thickness distribution. This process comprehensively considers the generation of bubbles in the current casting process and the unevenness of the thickness distribution of the insulation layer. If the bubble parameters and distortion degree deviate from the target range by a large amount, the configured adjustment step will be larger so as to quickly correct the parameters. When the bubble parameters and distortion degree gradually approach the target values, the adjustment step will gradually decrease to improve the optimization accuracy. During the adjustment process, each optimization is based on the current casting parameters. Combined with the configured adjustment step, the casting parameters are iteratively updated. Through multiple rounds of iterations, the casting parameters are continuously optimized so that they gradually approach the parameter combination that can achieve the best casting effect. The optimization goal is to make the first bubble parameter and the first distortion degree reach the expected values, so as to obtain the best casting fitness. When the optimization converges, the casting parameters finally determined are the optimal casting parameters. The casting of the transformer winding using the determined optimal casting parameters can minimize the generation of bubbles and ensure the thickness uniformity and overall quality of the insulation layer, thereby meeting the quality requirements of actual production.

[0034] Further, the present application provides configuring a casting adjustment step according to the first bubble parameter and the first distortion degree, adjusting the first casting parameter, continuing to optimize the casting parameter, obtaining the optimal casting parameter, and casting the transformer winding, including: Obtain a preset pouring adjustment step for adjusting pouring parameters; calculate the bubble ratio of the first bubble parameter to the bubble quantity threshold value; use the average of the bubble ratio and the first distortion degree, multiply by the preset pouring adjustment step, to obtain the pouring adjustment step; use the pouring adjustment step to adjust the first pouring parameter, continue to optimize the pouring parameter until convergence, output the pouring parameter with the maximum pouring fitness, obtain the optimal pouring parameter, and cast the transformer winding.

[0035] Preferably, in the process of optimizing the pouring parameters, the system terminal first obtains a preset pouring adjustment step size, which is the default basic value in the initial adjustment process, determined based on historical experience and expert decision-making, and is used to control the amplitude of parameter updates; then, the bubble ratio is calculated based on the first bubble parameter and the set bubble quantity threshold value, and the ratio reflects the degree of deviation of the bubbles in the current pouring process. In order to ensure that the adjustment step size can dynamically adapt to the current pouring effect, the system terminal combines the bubble ratio and the average of the first distortion to correct the preset step size. Specifically, when the bubble parameters and the distortion deviate greatly from the ideal range, the pouring effect is poor. At this time, the step size will be enlarged to jump out of the current poor pouring parameter area and quickly approach the potential excellent pouring parameters. On the contrary, when the bubble parameters and the distortion gradually approach the ideal value, it means that the pouring effect is good. At this time, the step size will be reduced to perform fine adjustment near the optimal parameters. , find possible better solutions, and obtain the pouring adjustment step size. The specific method is to multiply the mean of the bubble ratio and the distortion by the preset pouring adjustment step size to obtain the pouring adjustment step size required for the current iteration. Through this dynamic adjustment, the adaptive change of the step size can be achieved, which can not only quickly jump out of the inferior solution area, but also accurately optimize near the optimal solution; then, the current pouring parameters are updated using the calculated pouring adjustment step size. After the parameters are updated, the pouring effect is evaluated again, and the above process is repeated. Through multiple rounds of iterations, the pouring parameters are continuously optimized until the parameter adjustment converges, that is, the pouring fitness reaches the expected fitness. Finally, the pouring parameters with the largest pouring fitness are output as the optimal solution. The optimal pouring parameters obtained will be used in the actual transformer winding pouring process. This parameter can minimize the generation of bubbles while ensuring the uniform thickness of the insulation layer, thereby achieving the best pouring effect and meeting the requirements of high-quality production.

[0036] In summary, the embodiments of the present application have at least the following technical effects: The embodiment of the present application optimizes the casting quality by constructing a three-dimensional model of the winding and a casting parameter space, combining bubble prediction and an insulation layer thickness distribution generation method, and calculating the casting fitness through the distortion of the bubble parameters and the insulation layer thickness distribution, and using a generative adversarial network to predict the thickness distribution, and then dynamically adjust the step size of the casting parameters, continuously optimize the casting process, and finally obtain the optimal casting parameters, ensure the uniformity and quality consistency of the insulation layer, and improve the casting efficiency and product performance. These technical effects jointly solve the technical problems of bubble generation and uneven insulation layer thickness during the casting process of the transformer winding, and achieve the technical effect of avoiding the generation of bubbles and improving the casting quality and performance of the transformer winding by dynamically adjusting the casting process parameters.

[0037] Embodiment 2, based on the same inventive concept as the method for optimizing the casting of transformer windings in the above embodiment, Figure 2 As shown, the present application provides a casting optimization system for transformer windings, and the system includes: a parameter space acquisition module 1: constructing a three-dimensional winding model of a transformer winding to be cast, and acquiring a casting parameter space; a parameter distribution acquisition module 2: randomly generating a first casting parameter in the casting parameter space, and combining the three-dimensional winding model to perform bubble prediction and insulation layer thickness distribution generation after casting, and obtain a first bubble parameter and a first insulation layer thickness distribution; a parameter distribution analysis module 3: performing a distortion recognition analysis on the first insulation layer thickness distribution, obtaining a first distortion degree, and calculating a first casting fitness in combination with the first bubble parameter; a transformer winding casting module 4: configuring a casting adjustment step according to the first bubble parameter and the first distortion degree, adjusting the first casting parameter, continuing to optimize the casting parameter, obtaining the optimal casting parameter, and casting the transformer winding.

[0038] Furthermore, the parameter space acquisition module 1 is also used to execute the following method: The surface morphology of the transformer winding to be cast is collected to construct a three-dimensional winding model of the transformer winding; the casting speed adjustment space for casting the transformer winding is obtained and constructed as a casting parameter space.

[0039] Furthermore, the parameter distribution obtaining module 2 is also used to perform the following method: A first casting parameter is randomly generated in the casting parameter space; the first casting parameter and the winding three-dimensional model are input into a pre-constructed bubble prediction path to predict and obtain the first bubble parameter, wherein the bubble prediction path is constructed by training with sample casting parameters, a sample winding three-dimensional model set and a sample bubble parameter set, and each sample bubble parameter includes the number of sample bubbles; according to the winding three-dimensional model and the first casting parameter, the thickness distribution of the insulation layer after casting is generated to obtain the first insulation layer thickness distribution.

[0040] Furthermore, the parameter distribution obtaining module 2 is also used to perform the following method: According to the transformer winding casting detection data in historical time, a sample casting parameter set and a sample winding three-dimensional model set are collected, and the insulation layer thicknesses at multiple positions of the transformer winding after casting with different sample casting parameters and sample winding three-dimensional models are collected to obtain a sample insulation layer thickness distribution set; based on a generative adversarial network, an insulation layer thickness generation path is constructed, wherein the insulation layer thickness generation path includes a generator and a discriminator; the sample casting parameter set, the sample winding three-dimensional model set and the sample insulation layer thickness distribution set are used as supervised training data to train the generator and the discriminator until convergence to obtain an insulation layer thickness generation path; the winding three-dimensional model and the first casting parameter are input into the insulation layer thickness generation path to generate a first insulation layer thickness distribution.

[0041] Furthermore, the parameter distribution analysis module 3 is also used to perform the following method: According to the winding three-dimensional model, the maximum depth difference between the recessed area and the raised area of ​​the transformer winding is extracted; according to the first insulating layer thickness distribution, the difference between the minimum insulating layer thickness and the maximum insulating layer thickness is calculated to obtain the first insulating layer thickness difference; the ratio of the first insulating layer thickness difference to the maximum depth difference is calculated as the first distortion degree; according to the first distortion degree and the first bubble parameter, the first casting adaptability of the first casting parameter is calculated.

[0042] Furthermore, the parameter distribution analysis module 3 is also used to perform the following method: According to the first distortion degree and the first bubble parameter, the first pouring adaptability of the first pouring parameter is calculated and obtained as follows: ; Among them, JZF is the casting adaptability, is the bubble weight, is the distortion weight, and The sum of is 1, is the bubble parameter, is the bubble number threshold value of transformer winding casting, K is the thickness variance of the first insulation layer thickness distribution, and J is the distortion degree.

[0043] Furthermore, the transformer winding casting module 4 is also used to perform the following method: Obtain a preset pouring adjustment step for adjusting pouring parameters; calculate the bubble ratio of the first bubble parameter to the bubble quantity threshold value; use the average of the bubble ratio and the first distortion degree, multiply by the preset pouring adjustment step, to obtain the pouring adjustment step; use the pouring adjustment step to adjust the first pouring parameter, continue to optimize the pouring parameter until convergence, output the pouring parameter with the maximum pouring fitness, obtain the optimal pouring parameter, and cast the transformer winding.

[0044] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0045] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0046] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for optimizing the casting of transformer windings, characterized in that: The method comprises: Constructing a three-dimensional winding model of a transformer winding to be cast, and obtaining a casting parameter space; Randomly generate a first casting parameter in the casting parameter space, and perform bubble prediction and insulation layer thickness distribution generation after casting in combination with the winding three-dimensional model to obtain a first bubble parameter and a first insulation layer thickness distribution; Performing a distortion recognition analysis on the thickness distribution of the first insulating layer to obtain a first distortion degree, and calculating a first casting adaptability in combination with the first bubble parameter; The casting adjustment step is configured according to the first bubble parameter and the first distortion degree, the first casting parameter is adjusted, the casting parameter is continuously optimized, the optimal casting parameter is obtained, and the casting of the transformer winding is performed.

2. The method for optimizing the casting of transformer windings according to claim 1, characterized in that: Construct a three-dimensional winding model of the transformer winding to be cast, and obtain the casting parameter space, including: Collecting the surface morphology of the transformer winding to be cast, and constructing a three-dimensional winding model of the transformer winding; The pouring speed adjustment space for pouring the transformer winding is obtained and constructed as a pouring parameter space.

3. The method for optimizing the casting of transformer windings according to claim 1, characterized in that: Randomly generating a first casting parameter in the casting parameter space, combining the winding three-dimensional model, performing bubble prediction after casting and generating the insulation layer thickness distribution, and obtaining the first bubble parameter and the first insulation layer thickness distribution, including: Randomly generating a first pouring parameter in the pouring parameter space; Inputting the first casting parameter and the winding three-dimensional model into a pre-constructed bubble prediction path to predict and obtain a first bubble parameter, wherein the bubble prediction path is obtained by training and constructing a sample casting parameter, a sample winding three-dimensional model set, and a sample bubble parameter set, and each sample bubble parameter includes the number of sample bubbles; According to the winding three-dimensional model and the first casting parameters, the thickness distribution of the insulation layer after casting is generated to obtain the first insulation layer thickness distribution.

4. The method for optimizing the casting of transformer windings according to claim 3, characterized in that: Generating a thickness distribution of the insulation layer after casting according to the winding three-dimensional model and the first casting parameter to obtain a first insulation layer thickness distribution includes: According to the transformer winding casting detection data in the historical time, a sample casting parameter set and a sample winding three-dimensional model set are collected, and the insulation layer thickness at multiple positions of the transformer winding after casting with different sample casting parameters and sample winding three-dimensional models is collected to obtain a sample insulation layer thickness distribution set; Based on a generative adversarial network, constructing an insulating layer thickness generation path, wherein the insulating layer thickness generation path includes a generator and a discriminator; The sample casting parameter set, the sample winding three-dimensional model set and the sample insulation layer thickness distribution set are used as supervised training data to train the generator and the discriminator until convergence, thereby obtaining an insulation layer thickness generation path; The winding three-dimensional model and the first casting parameter are input into the insulation layer thickness generation path to generate a first insulation layer thickness distribution.

5. The method for optimizing the casting of transformer windings according to claim 1, characterized in that: Performing a distortion recognition analysis on the thickness distribution of the first insulating layer to obtain a first distortion degree, and calculating and obtaining a first casting adaptability in combination with the first bubble parameter, including: Extracting the maximum depth difference between the concave area and the convex area of ​​the transformer winding according to the winding three-dimensional model; Calculating the difference between the minimum insulation layer thickness and the maximum insulation layer thickness according to the first insulation layer thickness distribution to obtain a first insulation layer thickness difference; Calculating a ratio of the first insulating layer thickness difference to the maximum depth difference as a first distortion degree; A first casting adaptability of a first casting parameter is obtained by calculation according to the first distortion degree and the first bubble parameter.

6. The method for optimizing the casting of transformer windings according to claim 5, characterized in that: According to the first distortion degree and the first bubble parameter, the first pouring adaptability of the first pouring parameter is calculated and obtained as follows: ； Among them, JZF is the casting adaptability, is the bubble weight, is the distortion weight, and The sum of is 1, is the bubble parameter, is the bubble number threshold value of transformer winding casting, K is the thickness variance of the first insulation layer thickness distribution, and J is the distortion degree.

7. The method for optimizing the casting of transformer windings according to claim 6, characterized in that: Configuring a casting adjustment step length according to the first bubble parameter and the first distortion degree, adjusting the first casting parameter, continuing to optimize the casting parameter, obtaining an optimal casting parameter, and casting the transformer winding, including: Get the preset pouring adjustment step length for adjusting pouring parameters; Calculate the bubble ratio of the first bubble parameter to the bubble quantity threshold value; The average of the bubble ratio and the first distortion degree is multiplied by the preset pouring adjustment step length to obtain the pouring adjustment step length; The first casting parameter is adjusted by using the casting adjustment step, and the casting parameter optimization is continued until convergence, and the casting parameter with the maximum casting fitness is output to obtain the optimal casting parameter, and the casting of the transformer winding is performed.

8. A casting optimization system for transformer windings, characterized in that: The steps for implementing the casting optimization method of a transformer winding as claimed in any one of claims 1 to 7 include: Parameter space acquisition module: constructs a three-dimensional winding model of the transformer winding to be cast and obtains the casting parameter space; Parameter distribution acquisition module: randomly generates a first casting parameter in the casting parameter space, and performs bubble prediction and insulation layer thickness distribution generation after casting in combination with the winding three-dimensional model to obtain a first bubble parameter and a first insulation layer thickness distribution; Parameter distribution analysis module: performs distortion recognition analysis on the thickness distribution of the first insulating layer to obtain a first distortion degree, and calculates a first pouring adaptability in combination with the first bubble parameter; Transformer winding casting module: configure the casting adjustment step according to the first bubble parameter and the first distortion degree, adjust the first casting parameter, continue to optimize the casting parameter, obtain the optimal casting parameter, and cast the transformer winding.

Images