A standardized power module design method and system for different application scenarios

Abstract

The present invention provides a standardized power module design method and system for different application scenarios, the method is to generate several reference module design schemes by clustering analysis of power module parameter index data of multiple target scenarios, and to generate filter manufacturing process files by lightweight optimization design of passive filters in the reference module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, and to generate fault protection circuit design files by performing module fault simulation analysis on each reference module design scheme respectively, and to generate the optimal standardized module design scheme for the scenario based on the preset module performance index system according to the reference module design scheme, filter manufacturing process files and fault protection circuit design files. The present invention can not only improve the design efficiency of power modules in different application scenarios, but also effectively balance the application requirements of standardization, lightweight and high reliability, and can also improve the environmental adaptability of power modules.

Description

Technical Field

[0001] The present invention relates to the technical field of power module design, and in particular to a standardized power module design method and system for different application scenarios. Background Art

[0002] As the core component of the power electronic converter, the power module is a complete power conversion unit that integrates power devices, power device drivers, capacitors, filters, heat sinks and other components, and is mainly used in medium and high power scenarios. Faced with the differentiated requirements for the electrical performance, reliability and environmental adaptability of power modules in different application scenarios, how to achieve the standardization of scene power modules on the basis of ensuring performance has become a major problem in the design of power modules; at the same time, considering that the passive filter components inside the power module usually occupy a large volume, which is not conducive to improving the module integration, and the power module may encounter various fault conditions in actual operation, which will seriously threaten the safety of the module itself and its power supply object, how to reduce the size and weight of the filter components to achieve lightweight design of the module while ensuring filtering performance, and how to effectively improve the fault detection and protection capabilities of the power module to ensure rapid blocking and fault crossing have also become the focus of module design.

[0003] However, in actual applications, there are mutual constraints between the above-mentioned power module design issues. Most of the existing power module design solutions only focus on the implementation structure design of the required module functions in different application scenarios, or only conduct design research based on the above-mentioned design problem. They do not pay attention to the coordination and balance of the standardization, lightweight and high reliability of the power module during the design. This will inevitably lead to limited application effects of the power module and make it difficult to truly meet the usage needs in actual scenarios. Summary of the invention

[0004] The purpose of the present invention is to provide a standardized power module design method for different application scenarios. On the basis of classifying the power module parameter index data of different application scenarios to obtain a benchmark power module design scheme, a lightweight optimization design of the passive filter and an intelligent fault protection circuit design are performed. The design scheme is comprehensively evaluated and optimized based on a preset module performance index system that simultaneously considers multiple module characteristics such as standardization, lightweight and high reliability, and the optimal design scheme under different application scenarios is quickly generated. This can not only improve the design efficiency of the power module under different application scenarios, but also effectively balance the application requirements of standardization, lightweight and high reliability, and can also improve the environmental adaptability of the power module.

[0005] In order to achieve the above objectives, it is necessary to provide a standardized power module design method and system for different application scenarios in response to the above technical problems.

[0006] In a first aspect, an embodiment of the present invention provides a standardized power module design method for different application scenarios, wherein the standardized power module includes a passive filter device and a fault protection circuit; the method includes the following steps:

[0007] Obtain power module parameter index data for multiple target scenarios, perform cluster analysis on all power module parameter index data, and generate several benchmark module design solutions;

[0008] According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding benchmark module design solution is lightweight optimized and the corresponding filter manufacturing process file is generated;

[0009] Perform module fault simulation analysis on each benchmark module design scheme and generate corresponding fault protection circuit design files;

[0010] According to the benchmark module design scheme of each target scenario, the filter manufacturing process file and the fault protection circuit design file, the optimal standardized module design scheme for the corresponding scenario is generated based on the preset module performance indicator system.

[0011] Furthermore, the power module parameter index data includes electrical performance index, reliability index and environmental adaptability index;

[0012] The steps of performing cluster analysis on the electrical performance indicators, reliability indicators and environmental adaptability indicators under all target scenarios to generate several reference module design solutions include:

[0013] Preprocess the power module parameter index data under all target scenarios to obtain the corresponding standardized parameter index data set;

[0014] According to the standardized parameter indicator data set, key parameter indicator screening is performed based on correlation analysis and principal component analysis to obtain a corresponding key parameter indicator data set;

[0015] K-means clustering analysis is performed on the key parameter indicator data set to obtain several benchmark module design solutions.

[0016] Furthermore, after the step of performing cluster analysis on all power module parameter index data to generate a plurality of reference module design solutions, the method further includes:

[0017] According to each benchmark module design scheme, establish the corresponding power module simulation model;

[0018] Perform multi-physics field simulation analysis on each power module simulation model to obtain the corresponding module performance evaluation results;

[0019] According to the module performance evaluation results, the structural parameters and topological parameters of the corresponding power module simulation model are optimized to generate a module selection guidance strategy;

[0020] Abstracting each reference module design scheme based on parametric modeling technology to generate a corresponding module parametric model; the module parametric model includes structural parameters, device parameters and material parameters;

[0021] Each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model are stored in a preset relational database to generate a corresponding power module knowledge base.

[0022] Furthermore, the step of performing lightweight optimization design on the passive filter in the corresponding reference module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario and generating the corresponding filter manufacturing process file includes:

[0023] According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structure model of the corresponding passive filter is established;

[0024] According to the initial structural model, a component material property data set, a target volume range, and a target weight range of the corresponding passive filter are obtained; the component material property data set includes material data and shape data of various components;

[0025] Performing component size prediction analysis according to the component material property data set, the target volume range, the target weight range and a pre-constructed filter optimal size prediction model to generate a corresponding filter component optimal size set; the filter optimal size prediction model is obtained based on a preset feedforward neural network training;

[0026] According to the optimal size set of the filter element, the initial structural model is optimized to generate a corresponding first optimized structural model;

[0027] By using electromagnetic field simulation software, simulation analysis of electromagnetic field distribution and filtering performance of the first optimized structural model is performed to generate corresponding simulation analysis results;

[0028] According to the simulation analysis result, fine-tuning the corresponding first optimized structural model to generate a corresponding second optimized structural model;

[0029] The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; the filter manufacturing process file includes component materials, component size specifications and processing technology.

[0030] Furthermore, the steps of constructing the filter optimal size prediction model include:

[0031] Acquire a passive filter design parameter data set; the passive filter design parameter data set includes component material property data, component size data, and corresponding device volume weight data of the passive filter under different target scenarios;

[0032] The preset feedforward neural network is optimized and trained using component material property data and corresponding component size data as model input and model output respectively, and the corresponding device volume weight data as optimization constraints to obtain the filter optimal size prediction model.

[0033] Furthermore, the step of performing module fault simulation analysis on each reference module design scheme to generate a corresponding fault protection circuit design file includes:

[0034] According to the preset fault conditions, module fault simulation is performed on each benchmark module design scheme, and the corresponding module fault operation data set is collected;

[0035] Preprocessing the module fault operation data to generate a module status feature data set;

[0036] According to the module status feature data set, a support vector machine model is trained to construct a corresponding module fault detection model;

[0037] The fault protection circuit design file is generated according to the module fault detection model and the corresponding fault condition processing strategy.

[0038] Furthermore, the preset module performance index system includes a standardization degree index, a lightweight index and a reliability index;

[0039] The step of generating the optimal standardized module design scheme for the corresponding scenario based on the preset module performance indicator system according to the reference module design scheme for each target scenario, the filter manufacturing process file and the fault protection circuit design file comprises:

[0040] Taking the maximization of the module's continuous working power and energy utilization efficiency as the optimization goals, a corresponding scenario multi-objective optimization model is established based on the preset scenario voltage range and current range constraints of each target scenario;

[0041] By means of the hierarchical analysis method, the initial weight vector of the indicators corresponding to the preset module performance indicator system under each target scenario is obtained, and the fitness function is constructed according to the weighted average value of each module performance indicator in the preset module performance indicator system;

[0042] The reference module design scheme of each target scenario and the corresponding filter manufacturing process file and the fault protection circuit design file are combined and designed to generate the corresponding scenario initial module design scheme, and the module technical parameters and module performance index data corresponding to the scenario initial module design scheme are obtained; the module technical parameters include output voltage, output current and input power;

[0043] According to the module technical parameters, the module performance index data and the initial index weight vector of the initial module design scheme of each scenario, a genetic algorithm is executed based on the fitness function to solve the scenario multi-objective optimization model to obtain the corresponding scenario optimal index weight vector;

[0044] According to the optimal indicator weight vector of each scenario, the parameters of the initial module design scheme of the corresponding scenario are optimized to generate the optimal standardized module design scheme of the corresponding scenario.

[0045] Furthermore, the method further comprises:

[0046] Acquire environmental condition data for each target scenario, perform environmental adaptability simulation analysis on the optimal standardized module design scheme for the corresponding scenario based on the environmental condition data, obtain the module units to be optimized, and adjust the design parameters and / or material selection of the units to be optimized in the optimal standardized module design scheme for the scenario.

[0047] Furthermore, the method further comprises:

[0048] According to the scene-optimal standardized module design plan for each target scene, the corresponding scene power module is produced, and the scene power module is subjected to online performance testing according to the pre-built performance anomaly prediction model, and the module production process is optimized and adjusted according to the corresponding performance test results.

[0049] In a second aspect, an embodiment of the present invention provides a standardized power module design system for different application scenarios, wherein the standardized power module includes a passive filter device and a fault protection circuit; the system includes:

[0050] Standardized design module, used to obtain power module parameter index data for multiple target scenarios, and perform cluster analysis on all power module parameter index data to generate several benchmark module design solutions;

[0051] The lightweight design module is used to optimize the passive filter in the corresponding benchmark module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, and generate the corresponding filter manufacturing process file;

[0052] Reliability design module, used to perform module fault simulation analysis on each reference module design scheme and generate corresponding fault protection circuit design files;

[0053] The optimal solution generation module is used to generate the optimal standardized module design solution for the corresponding scenario based on the preset module performance indicator system according to the benchmark module design solution of each target scenario, the filter manufacturing process file and the fault protection circuit design file.

[0054] Furthermore, the power module parameter index data includes electrical performance index, reliability index and environmental adaptability index; the power module parameter index data under all target scenarios are clustered and analyzed to generate several reference module design schemes, including:

[0055] The electrical performance indicators, reliability indicators and environmental adaptability indicators in all target scenarios are preprocessed to obtain the corresponding standardized parameter indicator data sets;

[0056] According to the standardized parameter indicator data set, key parameter indicator screening is performed based on correlation analysis and principal component analysis to obtain a corresponding key parameter indicator data set;

[0057] K-means clustering analysis is performed on the key parameter indicator data set to obtain several benchmark module design solutions.

[0058] Furthermore, after clustering all power module parameter index data to generate a number of reference module design solutions, the method further includes:

[0059] According to each benchmark module design scheme, establish the corresponding power module simulation model;

[0060] Perform multi-physics field simulation analysis on each power module simulation model to obtain the corresponding module performance evaluation results;

[0061] According to the module performance evaluation results, the structural parameters and topological parameters of the corresponding power module simulation model are optimized to generate a module selection guidance strategy;

[0062] Abstracting each reference module design scheme based on parametric modeling technology to generate a corresponding module parametric model; the module parametric model includes structural parameters, device parameters and material parameters;

[0063] Each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model are stored in a preset relational database to generate a corresponding power module knowledge base.

[0064] Furthermore, according to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding reference module design scheme is lightweight optimized and designed to generate the corresponding filter manufacturing process file, including:

[0065] According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structure model of the corresponding passive filter is established;

[0066] According to the initial structural model, a component material property data set, a target volume range, and a target weight range of the corresponding passive filter are obtained; the component material property data set includes material data and shape data of various components;

[0067] Performing component size prediction analysis according to the component material property data set, the target volume range, the target weight range and a pre-constructed filter optimal size prediction model to generate a corresponding filter component optimal size set; the filter optimal size prediction model is obtained based on a preset feedforward neural network training;

[0068] According to the optimal size set of the filter element, the initial structural model is optimized to generate a corresponding first optimized structural model;

[0069] By using electromagnetic field simulation software, simulation analysis of electromagnetic field distribution and filtering performance of the first optimized structural model is performed to generate corresponding simulation analysis results;

[0070] According to the simulation analysis result, fine-tuning the corresponding first optimized structural model to generate a corresponding second optimized structural model;

[0071] The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; the filter manufacturing process file includes component materials, component size specifications and processing technology.

[0072] Furthermore, the construction of the filter optimal size prediction model includes:

[0073] Acquire a passive filter design parameter data set; the passive filter design parameter data set includes component material property data, component size data, and corresponding device volume weight data of the passive filter under different target scenarios;

[0074] The preset feedforward neural network is optimized and trained using component material property data and corresponding component size data as model input and model output respectively, and the corresponding device volume weight data as optimization constraints to obtain the filter optimal size prediction model.

[0075] Furthermore, the step of performing module fault simulation analysis on each reference module design scheme to generate a corresponding fault protection circuit design file includes:

[0076] According to the preset fault conditions, module fault simulation is performed on each benchmark module design scheme, and the corresponding module fault operation data set is collected;

[0077] Preprocessing the module fault operation data to generate a module status feature data set;

[0078] According to the module status feature data set, a support vector machine model is trained to construct a corresponding module fault detection model;

[0079] The fault protection circuit design file is generated according to the module fault detection model and the corresponding fault condition processing strategy.

[0080] Furthermore, the preset module performance index system includes a standardization degree index, a lightweight index and a reliability index; the reference module design scheme according to each target scenario, the filter manufacturing process file and the fault protection circuit design file, and the corresponding scenario optimal standardized module design scheme is generated based on the preset module performance index system, including:

[0081] Taking the maximization of the module's continuous working power and energy utilization efficiency as the optimization goals, a corresponding scenario multi-objective optimization model is established based on the preset scenario voltage range and current range constraints of each target scenario;

[0082] By means of the hierarchical analysis method, the initial weight vector of the indicators corresponding to the preset module performance indicator system under each target scenario is obtained, and the fitness function is constructed according to the weighted average value of each module performance indicator in the preset module performance indicator system;

[0083] The reference module design scheme of each target scenario and the corresponding filter manufacturing process file and the fault protection circuit design file are combined and designed to generate the corresponding scenario initial module design scheme, and the module technical parameters and module performance index data corresponding to the scenario initial module design scheme are obtained; the module technical parameters include output voltage, output current and input power;

[0084] According to the module technical parameters, the module performance index data and the initial index weight vector of the initial module design scheme of each scenario, a genetic algorithm is executed based on the fitness function to solve the scenario multi-objective optimization model to obtain the corresponding scenario optimal index weight vector;

[0085] According to the optimal indicator weight vector of each scenario, the parameters of the initial module design scheme of the corresponding scenario are optimized to generate the optimal standardized module design scheme of the corresponding scenario.

[0086] Furthermore, the system further comprises:

[0087] The adaptive optimization module is used to obtain the environmental condition data of each target scene, perform environmental adaptability simulation analysis on the optimal standardized module design scheme for the corresponding scene based on the environmental condition data, obtain the module units to be optimized, and adjust the design parameters and / or material selection of the units to be optimized in the optimal standardized module design scheme for the scene.

[0088] Furthermore, the system further comprises:

[0089] The test optimization module is used to produce corresponding scenario power modules according to the scenario optimal standardized module design scheme of each target scenario, and to perform online performance detection on the scenario power module according to the pre-built performance anomaly prediction model, and to optimize and adjust the module production process according to the corresponding performance detection results.

[0090] In a third aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0091] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0092] The present invention provides a standardized power module design method, system, computer equipment and storage medium for different application scenarios. The method can achieve the acquisition of power module parameter index data for multiple target scenarios, perform cluster analysis on all power module parameter index data to generate several benchmark module design schemes, perform lightweight optimization design on the passive filter in the corresponding benchmark module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, generate the corresponding filter manufacturing process file, perform module fault simulation analysis on each benchmark module design scheme, generate the corresponding fault protection circuit design file, and generate the technical solution of the optimal standardized module design scheme for the corresponding scenario based on the preset module performance index system according to the benchmark module design scheme of each target scenario, the filter manufacturing process file and the fault protection circuit design file. Compared with the existing technology, this standardized power module design method for different application scenarios classifies the power module parameter index data of different application scenarios to obtain a benchmark power module design scheme, and then performs lightweight optimization design of the passive filter and intelligent fault protection circuit design. The design scheme is comprehensively evaluated and optimized based on a preset module performance index system that simultaneously considers multiple module characteristics such as standardization, lightweight and high reliability, and the optimal design scheme for different application scenarios is quickly generated. This can not only improve the design efficiency of the power module in different application scenarios, but also effectively balance the application requirements of standardization, lightweight and high reliability, and can also improve the environmental adaptability of the power module, and has high application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 It is a flow chart of a standardized power module design method for different application scenarios in an embodiment of the present invention;

[0094] Figure 2 It is a structural schematic diagram of a standardized power module design system for different application scenarios in an embodiment of the present invention;

[0095] Figure 3 It is a diagram of the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0096] In order to make the purpose, technical scheme and beneficial effects of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Obviously, the embodiments described below are part of the embodiments of the present invention and are only used to illustrate the present invention, but are not used to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0097] The standardized power module design method for different application scenarios provided by the present invention can be understood as a design optimization scheme that effectively balances the multiple characteristics of standardization, lightweight, reliability, etc. of power module design for different application scenarios, based on the current situation that the existing power module design does not consider the application requirements of balancing standardization, lightweight, and high reliability, resulting in limited actual application effects. The following embodiments will explain the standardized power module design method for different application scenarios of the present invention in detail.

[0098] In one embodiment, Figure 1 As shown, a standardized power module design method for different application scenarios is provided, wherein the standardized power module includes a passive filter device and a fault protection circuit; and the following steps are included:

[0099] S11. Obtain power module parameter index data for multiple target scenarios, and perform cluster analysis on all power module parameter index data to generate several benchmark module design schemes; wherein, the target scenario can be understood as the power scenario required for the actual power module, which varies according to the actual application requirements and is not specifically limited here. The power module parameter index data can be understood as parameter data that can reflect the design requirements of the power module under different application scenarios. In order to facilitate the distinction of design requirements for different application scenarios, this embodiment preferably sets the power module parameter index data to include electrical performance indicators, reliability indicators, and environmental adaptability indicators, and the electrical performance indicators include input voltage range, output voltage, and output current, etc., the reliability indicators include mean time between failures and failure rate, etc., and the environmental adaptability indicators include operating temperature range, relative humidity range, etc.

[0100] The reference module design scheme in this embodiment can be understood as a standard design scheme of a power module that can specifically meet the functional requirements of a certain type of application scenario and includes a variety of standard components, including basic design methods such as power device selection, circuit layout and heat dissipation design that can be equivalently obtained through the above power module parameter index data. Since there are certain differences in the structural design of power modules in different application scenarios, the specific contents of the reference module design scheme in this embodiment are not limited here; specifically, the steps of clustering the power module parameter index data under all target scenarios to generate several reference module design schemes include:

[0101] The electrical performance indicators, reliability indicators and environmental adaptability indicators in all target scenarios are preprocessed respectively to obtain the corresponding standardized parameter indicator data sets; among them, the preprocessing can be understood as first scaling the data of each parameter indicator to the range of [0,1] through the minimum-maximum normalization method, and then removing the noise and outliers in the data through median filtering, which will not be described in detail here.

[0102] According to the standardized parameter indicator data set, key parameter indicators are screened based on correlation analysis and principal component analysis to obtain a corresponding key parameter indicator data set; wherein the key parameter indicator data set can be understood as a representative and discriminative key feature parameter data set obtained by screening the standardized parameter indicator data set. The specific acquisition process is to first use correlation analysis to perform a preliminary screening of the indicator parameters in the standardized parameter indicator data set, eliminate indicator parameters with low relevance to the application scenario, and then perform principal component analysis on the screened related standardized parameter indicator data, thereby reducing the data dimension while effectively retaining the most important data information, providing reliable guarantee for subsequent efficient and reliable clustering analysis.

[0103] K-means clustering analysis is performed on the key parameter indicator data set to obtain several benchmark module design schemes; wherein, the process of obtaining the benchmark module design scheme can be understood as after selecting clustering parameters according to the actual number of target scenarios, each cluster center is iteratively optimized through K-means clustering analysis, so that similar power module design schemes are divided into a corresponding number of benchmark design schemes and each benchmark design scheme can meet the usage requirements of a class of application scenarios, that is, the basic standardized design of power modules suitable for different application scenarios is realized.

[0104] At the same time, considering that the several benchmark module design solutions corresponding to the target scenarios obtained by the above method can be used as the inheritance of relevant module design knowledge and experience, this embodiment preferably further analyzes the benchmark module design solution to extract the corresponding parameterized model and selection guidance strategy to build a relevant knowledge base to provide a reference basis for subsequent power module design. Specifically, after the step of clustering all power module parameter index data to generate several benchmark module design solutions, it also includes:

[0105] According to each benchmark module design scheme, a corresponding power module simulation model is established; wherein, the power module simulation model can be understood as a circuit model constructed based on each benchmark module design scheme using the MATLAB simulation tool. For specific information on how to build a power module simulation model based on the MATLAB simulation tool, please refer to the relevant existing technology implementation, which will not be described in detail here.

[0106] Multi-physics field simulation analysis is performed on each power module simulation model to obtain the corresponding module performance evaluation results. Among them, multi-physics field simulation analysis can be understood as the interaction between multiple physical fields such as electromagnetic field, temperature field and structural field. By analyzing the current distribution, voltage distribution, electromagnetic loss, temperature rise of the power module under different working conditions, as well as the structural deformation and fatigue life under the action of electromagnetic force and thermal stress, the corresponding module performance evaluation results including efficiency, power density and electromagnetic compatibility data can provide an effective basis for the structural optimization, efficiency and reliability improvement of the power module.

[0107] According to the module performance evaluation results, the structural parameters and topological parameters of the corresponding power module simulation model are optimized to generate a module selection guidance strategy; wherein, the structural parameter and topological parameter optimization can be understood as adjusting and improving the circuit parameter settings and layout wiring methods based on the module performance evaluation results to meet the usage requirements of different application scenarios, and generating a power module selection guidance strategy suitable for communications, industry, automobiles and other fields to support the provision of recommended input and output parameter ranges, key component selection and other guidance information according to the needs of different application scenarios.

[0108] Based on the parametric modeling technology, each benchmark module design scheme is abstracted to generate a corresponding module parametric model; wherein, the module parametric model includes structural parameters, device parameters and material parameters; the corresponding parametric modeling technology can be understood as a rule-based parametric design technology, which abstracts the circuit topology, key components and material parameters of each benchmark module design scheme into a parametric model, and matches the corresponding design rules for the corresponding parametric model. In subsequent use, the specific circuit schematic and PCB design scheme can be automatically generated directly by associating the design rules with the parametric model.

[0109] Each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model are stored in a preset relational database to generate a corresponding power module knowledge base; wherein, the power module knowledge base can be understood as a relational database that maps and associates and stores each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model. The relationship between the three types of information can be described by using ontology technology and semantic web technology, which facilitates intelligent retrieval and reasoning based on the power module knowledge base, and provides a reference basis for the subsequent efficient and reliable design of power modules for different application scenarios.

[0110] S12. According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding benchmark module design solution is lightweight optimized and designed, and the corresponding filter manufacturing process file is generated; wherein, the lightweight optimization design of the passive filter can be understood as optimizing the material, shape, size and other parameters of the capacitors, inductors and other components in the passive filter under the premise of ensuring the filtering performance, thereby greatly reducing the volume and weight of the passive filter. The design optimization is convenient for improving the integration of the power module and realizing the miniaturization design of the power module. Specifically, according to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding benchmark module design solution is lightweight optimized and the steps of generating the corresponding filter manufacturing process file include:

[0111] According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structural model of the corresponding passive filter is established; wherein, the module electrical parameter requirements and filtering performance requirements vary according to the application scenario (target scenario) of the actual power module, and are not specifically limited here; the initial structural model of the passive filter can be understood as the basic structure of the passive filter that meets the module electrical parameter requirements and filtering performance requirements of the corresponding target scenario, including the component types such as inductance (L), capacitance (C) and resistance (R), material properties and connection relationships between components.

[0112] According to the initial structural model, a component material property data set, a target volume range and a target weight range of the corresponding passive filter are obtained; wherein the component material property data set includes material data and shape data of various components; the target volume range and the target weight range can be understood as the expected volume range and the expected weight range of the passive filter pre-set based on the lightweight effect of the preset power module, which can be set according to the actual application scenario and application requirements. For example, for a step-down switching power supply with an input voltage of 24V, an output voltage of 12V, and an output current of 5A, the output ripple is required to be less than 50mV, then the corresponding capacitor can be a rectangular X7R ceramic capacitor and the initial capacitance value is set to 100uF, the inductor can be a ring-shaped ferrite core and the initial inductance value is set to 10uH, the corresponding device target volume range can be set to less than 10cm³ and the target weight range can be set to less than 50g.

[0113] According to the component material property data set, the target volume range, the target weight range and the pre-constructed filter optimal size prediction model, the component size prediction analysis is performed to generate the corresponding filter component optimal size set; wherein, the filter optimal size prediction model can be understood as a network model that can predict and analyze the optimal design size of each component inside the passive filter based on the component material property data set, the target volume range and the target weight range of the passive filter. In order to facilitate the simple and efficient prediction and analysis of the optimal size of the filter components, this embodiment preferably adopts the filter optimal size prediction model obtained by training based on a preset feedforward neural network; specifically, the construction steps of the filter optimal size prediction model include:

[0114] Obtain a passive filter design parameter data set; wherein the passive filter design parameter data set can be understood as a related design parameter data set obtained based on the existing passive filter collection; specifically, the passive filter design parameter data set includes component material property data, component size data and corresponding device volume weight data of the passive filter under different target scenarios, wherein the component material property data includes component materials and corresponding material properties, such as the dielectric constant and loss factor of capacitors, the magnetic permeability of inductors, etc., which can be obtained through a pre-constructed property database of various component materials.

[0115] The preset feedforward neural network is optimized and trained with the component material property data and the corresponding component size data as the model input and model output respectively, and the corresponding device volume weight data is used as the optimization constraint to obtain the filter optimal size prediction model. It should be noted that in order to improve the prediction and analysis effect, a multi-layer feedforward neural network training can be used to construct the filter optimal size prediction model, and the input layer is set to receive the component material property data marked with component size data, the hidden layer is set to 20 neurons, and the output layer outputs the size parameters of each component. In the actual training process, the device volume weight data is used as a constraint condition (the loss function can be set accordingly or used in other ways, without specific limitation), and the neuron weights and thresholds in the network model are iteratively updated by dynamically adjusting the learning rate and regularization parameters based on the back propagation algorithm to avoid fitting or underfitting, until the model output result can meet the filter performance requirements, and the required filter optimal size prediction model is obtained.

[0116] In practical applications, the optimal size set of the required filter components can be obtained by inputting the component material property data set, target volume range, and target weight range under the corresponding target scenario into the pre-built filter optimal size prediction model for component size prediction analysis. For example, by using the filter optimal size prediction model to analyze the component material property data of the passive filter in the aforementioned step-down switching power supply that must meet the conditions of device volume less than 10cm³ and weight less than 50g, the optimal size data of the capacitor can be obtained, such as 5mm×5mm×2mm, and the inductor has a size of 6mm×6mm×4mm. It should be noted that the optimal size of the components given here is only an exemplary description.

[0117] According to the optimal size set of the filter elements, the initial structural model is optimized to generate a corresponding first optimized structural model; wherein the first optimized structural model can be understood as a passive filter structural model obtained by adjusting the element structure size in the initial structural model based on the optimal size of each element in the optimal size set of the filter elements.

[0118] Through electromagnetic field simulation software, the electromagnetic field distribution and filtering performance of the first optimized structural model are simulated and analyzed to generate corresponding simulation analysis results; wherein, the electromagnetic field simulation software can use ANSYS software, that is, ANSYS software is used to perform electromagnetic field simulation analysis on the optimized first optimized structural model, and corresponding simulation analysis results are obtained to determine whether the optimized structure can meet actual application requirements.

[0119] According to the simulation analysis results, the corresponding first optimized structural model is fine-tuned to generate the corresponding second optimized structural model; wherein the second optimized structural model can be understood as a filter structure obtained by further optimizing the structural parameters of the first optimized structural model according to the simulation analysis results after the simulation analysis results obtained by performing electromagnetic field simulation analysis on the first optimized structural model do not meet the actual filtering performance requirements. It should be noted that which structural parameters in the first optimized structural model are adjusted according to the simulation analysis results can be determined according to the circumstances and are not specifically limited here.

[0120] The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; wherein the filter manufacturing process file includes component materials, component size specifications and processing technology, that is, manufacturing materials, process flow, packaging form, etc. of components such as capacitors and inductors, which are used to guide the production and manufacturing of passive filter components and help realize the miniaturized design of power modules.

[0121] This embodiment optimizes the size and structure of components such as capacitors and inductors for passive filter devices inside the power module based on artificial neural networks. It can significantly reduce the volume and weight of the passive filter while ensuring the filtering performance, and provide reliable technical support for the lightweight and miniaturized design of the power module.

[0122] S13. Perform module fault simulation analysis on each benchmark module design scheme respectively, and generate corresponding fault protection circuit design files; wherein, module fault simulation analysis can be understood as performing fault simulation on the corresponding benchmark module design scheme for possible fault operating conditions under each target scenario, so as to collect power module operation data under different fault types for constructing the fault simulation and data collection and analysis process of the intelligent fault protection circuit deployed inside the module, so as to ensure the specific ability of the module to quickly block and cross faults, and improve the reliability of the power module. Correspondingly, the fault protection circuit design file can be understood as the design guidance file of the fault protection circuit in the power module, so as to ensure that the design can accurately identify and judge the fault type based on the actual operation data of the power module, and execute different protection strategies based on different fault types to ensure that the power module works in a safe state.

[0123] Specifically, the steps of performing module fault simulation analysis on each reference module design scheme and generating corresponding fault protection circuit design files include:

[0124] According to the preset fault conditions, module fault simulation is performed on each benchmark module design scheme, and the corresponding module fault operation data set is collected; wherein, the preset fault conditions vary according to the actual power module application scenarios and are not specifically limited here; the module fault simulation can be implemented with reference to the relevant existing power module fault simulation technology, and the corresponding collected module fault operation data set can be understood as data information that can truly reflect the module operation status, including signal data such as voltage and current.

[0125] The module fault operation data is preprocessed to generate a module status feature data set, wherein the module status feature data set can be understood as a preprocessed data set that meets the subsequent machine learning usage requirements after the module fault operation data is processed for outliers and missing values ​​in turn and then normalized. The specific acquisition process will not be described in detail here.

[0126] According to the module status feature data set, the support vector machine model is trained to construct a corresponding module fault detection model; wherein, the module fault detection model can be understood as a support vector machine model that can analyze and judge whether the operation is normal based on the model operation data collected in real time during the operation of the actual power module, as well as the corresponding fault type under abnormal operation conditions, and can provide a reliable basis for timely and reliable execution of fault protection during the operation of the power module. For example, in actual applications, if the voltage suddenly drops and the current rises sharply, the module fault detection model may determine that it is a short circuit fault, that is, it is in a relatively serious AC / DC short circuit fault state, and the protection program can be immediately started to quickly cut off the fault circuit through the protection circuit to prevent the fault from spreading. At the same time, the working state of the power module can be adjusted by control to reduce the power output to a safe level, such as reducing the power output to 50% of the standard operating level to protect the power module from being damaged by overload. In addition, after cutting off the faulty circuit, the system will not restore the circuit immediately, but will continue to monitor the status of the fault point based on the module fault detection model to confirm whether the fault has been completely isolated. Only when the voltage and current at the fault point are monitored to return to the normal range and run stably for a period of time, will the system consider gradually restoring the power output and eventually reconnecting the circuit to the system. The use of intelligent fault protection circuits that deploy module fault detection models can effectively improve the reliability and safety of power modules and ensure the stable operation of the power system. It should be noted that the training and construction process of the module fault detection model can refer to the training technology implementation of the relevant support vector machine model, which will not be repeated here.

[0127] The fault protection circuit design file is generated according to the module fault detection model and the corresponding fault condition handling strategy; wherein the fault condition handling strategy can be understood as a fault response strategy set in advance based on various fault types that may be output by the module fault detection model, combined with the operation and maintenance experience or a priori knowledge base of the actual application scenario. The specific fault condition handling strategy varies depending on the fault condition of the actual power module application scenario and is not specifically limited here.

[0128] This embodiment introduces a machine learning algorithm to design an intelligent fault protection circuit inside the power module, which can realize timely and accurate perception of fault types under different working conditions and quickly execute corresponding fault protection strategies, ensuring that the power module can quickly block and cross various possible faults, thereby effectively improving the operating reliability of the power module.

[0129] S14. Generate the optimal standardized module design scheme for the corresponding scenario based on the preset module performance index system according to the benchmark module design scheme of each target scenario, the filter manufacturing process file and the fault protection circuit design file; wherein, the preset module performance index system can be understood as a comprehensive module performance evaluation index designed based on the standardization, lightweight and reliability of the power module. Specifically, the preset module performance index system includes a standardization degree index, a lightweight index and a reliability index. Among them, the standardization degree can be quantified by the ratio of the number of standard parts in the module to the total number of parts, with a value range of 0 to 1; the lightweight index is quantified by the actual weight or volume of the module, and can be directly expressed in kg; the reliability index can be quantified by the mean time between failures (MTBF), in hours.

[0130] The scenario-optimal standardized module design scheme in this embodiment can be understood as the optimal module design scheme that can effectively balance the application requirements of standardization, lightweight and reliability by coordinating and optimizing the benchmark module design scheme, filter manufacturing process file and fault protection circuit design file under each target scenario based on each module performance index in the above-mentioned preset module performance index system. Specifically, the steps of generating the corresponding scenario-optimal standardized module design scheme based on the preset module performance index system according to the benchmark module design scheme of each target scenario, the filter manufacturing process file and the fault protection circuit design file include:

[0131] Taking the maximization of the continuous working power of the module and the maximization of the energy utilization efficiency as the optimization goals, the corresponding scenario multi-objective optimization model is established based on the preset scenario voltage range and current range constraints of each target scenario; wherein the objective function of the scenario multi-objective optimization model can simultaneously include the objective function of maximizing the continuous working power of the module and the objective function of maximizing the energy utilization efficiency, which are respectively expressed as:

[0132]

[0133]

[0134] In the formula,

[0135]

[0136] in, and They represent the objective function of maximizing the continuous working power of the module and the objective function of maximizing the energy utilization efficiency respectively; and Respectively represent the output voltage and output current of the power module; and Respectively represent the input power and power loss of the power module.

[0137] Through the hierarchical analysis method, the initial weight vector of the indicators corresponding to the preset module performance indicator system under each target scenario is obtained, and the fitness function is constructed according to the weighted average value of each module performance indicator in the preset module performance indicator system; wherein, the initial weight vector of the indicators includes the initial weight value of each module performance indicator, and the process of determining the initial weight vector of the indicators through the hierarchical analysis method can refer to the relevant existing technology. The fitness function can be understood as the evaluation basis for selecting the optimal individual when the indicator weights of the preset module performance indicator system under each target scenario are iteratively optimized through the genetic algorithm, which is expressed as:

[0138]

[0139] in, represents the fitness function; , and They represent the standardization degree index, lightweight index and reliability index respectively; , and Indicates the indicator weight.

[0140] The benchmark module design scheme of each target scenario and the corresponding filter manufacturing process file and fault protection circuit design file are combined and designed to generate the corresponding scenario initial module design scheme, and the module technical parameters and module performance index data corresponding to the scenario initial module design scheme are obtained; the module technical parameters include output voltage, output current and input power; wherein, the scenario initial module design scheme can be understood as the scenario module design scheme to be optimized obtained by directly updating and replacing the relevant designs in the benchmark module design scheme based on the combined design of the filter manufacturing process file and the fault protection circuit design file, as the corresponding adjustment object for obtaining the scenario optimal index weight vector based on the following optimization algorithm.

[0141] According to the module technical parameters, the module performance indicator data and the initial weight vector of the indicators of the initial module design scheme of each scenario, the genetic algorithm is executed based on the fitness function to solve the scenario multi-objective optimization model to obtain the corresponding scenario optimal indicator weight vector; wherein, the module technical parameters can be used to calculate the objective function value in the scenario multi-objective optimization model, the module performance indicator data and the initial weight vector of the indicators can be used to calculate the fitness function value of each individual in the iterative optimization process of the genetic algorithm, and the corresponding update iterative formula of the scenario indicator weight vector involved in the process of iteratively solving the scenario multi-objective optimization model is based on the actual application scenario requirements, with reference to the existing genetic algorithm related design implementation, and will not be described in detail here.

[0142] According to the optimal index weight vector of each scenario, the corresponding scenario initial module design scheme is parameter optimized to generate the corresponding scenario optimal standardized module design scheme; wherein, the scenario optimal index weight vector can be understood as the optimal proportion of the corresponding various requirements in the power module design when the three design requirements of standardization, lightweight and high reliability can be effectively balanced. For example, if the initial weights corresponding to the standardization degree index, lightweight index and reliability index in a target scenario are 2, 3 and 5 respectively, and the optimal index weights obtained based on genetic algorithm optimization are 2, 2 and 6 respectively, then it is necessary to appropriately reduce the lightweight standard and improve the reliability design requirements, such as by increasing the working voltage to 400V, optimizing the heat dissipation structure or selecting high-reliability components to further optimize the reliability design of the power module, and after optimization, the finite element simulation analysis method is used to perform performance evaluation again, such as using thermal simulation analysis to determine whether the maximum temperature is lower than the maximum operating temperature of the components, and using vibration simulation analysis to determine whether the stress level is lower than the fatigue limit of the material, etc. If the performance evaluation result still does not meet the standard, then the optimization design is performed again until the best solution that meets the performance requirements is obtained as the final scenario optimal standardized module design scheme. Based on the optimal standardized module design solution for this scenario, the corresponding power module design parameters and process requirements can be generated, such as PCB wiring parameters, component packaging form, radiator installation method, etc., to guide product production and manufacturing and ensure the consistency and reliability of batch products.

[0143] This embodiment, on the basis of classifying the power module parameter index data of different application scenarios to obtain a benchmark power module design scheme, further introduces the lightweight optimization design of the passive filter and the intelligent fault protection circuit design, and comprehensively evaluates and optimizes the design scheme based on a preset module performance index system that simultaneously considers multiple module characteristics such as standardization, lightweight and high reliability, so as to quickly generate the optimal power module design scheme for different application scenarios. This can not only improve the design efficiency of the power module in different application scenarios, but also effectively balance the application requirements of standardization, lightweight and high reliability, thereby providing reliable guarantee for the operation effect of the power module, and has high application value.

[0144] Considering that the actual power module operation state is easily affected by environmental conditions, in order to improve the environmental adaptability of the power module and ensure the operating stability of the power model, this embodiment preferably obtains a module design solution that can effectively balance application requirements such as standardization, lightweight and high reliability, and further identifies the design weaknesses in the optimal standardized module design solution for each target scenario by introducing environmental adaptability simulation analysis and performs targeted adjustments and optimizations. Specifically, the method also includes:

[0145] Acquire environmental condition data for each target scenario, perform environmental adaptability simulation analysis on the optimal standardized module design scheme for the corresponding scenario based on the environmental condition data, obtain the module units to be optimized, and adjust the design parameters and / or material selection of the units to be optimized in the optimal standardized module design scheme for the scenario; wherein the environmental condition data include data such as temperature, humidity, and vibration conditions of the operating environment of the power module in each target scenario collected by sensors.

[0146] In practical applications, after obtaining the environmental condition data of each target environment, the corresponding power module simulation model can be built based on the optimal standardized module design scheme of the scenario of different target environments, and the power module simulation model can be subjected to finite element simulation operation analysis under the corresponding environmental condition data, and the power module can be evaluated and analyzed from the aspects of heat dissipation, insulation and protection. According to the evaluation and analysis results obtained, the design weak links of insufficient heat dissipation or poor insulation in the current power module are identified, and the identified design weak links are used as module optimization units that need further adjustment. Then, the design parameters or material selection are adjusted for each module optimization unit, and the adjusted design parameters or material selection are re-entered into the simulation software to evaluate the improved power module again, until a power module design scheme that meets the preset environmental adaptation requirements is obtained. For example, the ambient temperature in a certain target environment is 55°C, and the simulation analysis results show that when the ambient temperature exceeds 50°C, the heat dissipation performance of the power module decreases significantly. In order to address the problem of insufficient heat dissipation performance, the heat dissipation structure needs to be identified as the module unit to be optimized. This can be achieved by increasing the number of heat sinks or replacing the heat dissipation structure material with an aluminum alloy with high thermal conductivity. After the improvement, the simulation analysis is performed again under the same conditions. The results show that the temperature of the power module is 5°C lower than the original design, which proves the effectiveness of the optimized design.

[0147] This embodiment further optimizes and adjusts the scene-optimal standardized module design scheme under each target scenario based on environmental adaptability simulation analysis, so that the designed power module can be effectively adapted to the operating environment of the target scenario, effectively ensuring the operating performance and continuous stability of the power module, and thus meeting the strict requirements of actual industrial applications.

[0148] In addition, considering that the production process problems of power modules will also affect the application effect of power modules, in order to ensure the reliability of the power module production process corresponding to each target scenario, this embodiment preferably also performs online performance monitoring of the production scenario power modules by building a performance anomaly prediction model, timely discovers performance anomalies and potential faults introduced by production process problems, and optimizes and adjusts the relevant production processes. Specifically, the method also includes:

[0149] According to the optimal standardized module design scheme for each target scenario, the corresponding scenario power module is produced, and the scenario power module is subjected to online performance detection according to the pre-built performance anomaly prediction model, and the module production process is optimized and adjusted according to the corresponding performance detection results; it should be noted that the performance anomaly prediction model can be understood as a network model that is obtained by training a preset neural network model based on relevant data such as voltage, current and temperature under normal and abnormal performance conditions of the power module collected in the target scenario, which can accurately and efficiently identify performance anomalies or potential faults based on real-time collected voltage, current and temperature data. The specific network structure and type can be set according to the actual application and are not specifically limited here.

[0150] In practical applications, the deployed power module test system can be used to collect test data such as voltage, current, and temperature of the scene power modules obtained in production in real time, and the online test data can be preprocessed accordingly and input into the pre-built performance anomaly prediction model for performance anomaly prediction; if the performance anomaly prediction result is that there is a performance anomaly or potential fault, the correlation between the module test data and the production process-related parameters is further analyzed to locate the production process factors that cause the problem, and the production process is adjusted according to the analysis results, including but not limited to optimizing material ratios and adjusting process parameters, etc. The adjusted production process is applied to the production of new scene power modules, and the produced scene power modules are evaluated based on the performance anomaly prediction model until it is determined that the performance meets the standards to ensure that the corresponding production process defects are eliminated.

[0151] This embodiment further adopts a pre-built performance anomaly prediction model to conduct real-time online testing and analysis of the production scenario power modules, timely perceive the production process design defects and make reasonable optimization adjustments, and can achieve full-process optimization design of the power module from basic design to production process, providing reliable guarantee for the application performance of the power module.

[0152] At the same time, after the standardized, lightweight and highly reliable scene power modules that meet the application requirements of each target scene obtained through the above method steps are put into operation, it is also possible to build a remote monitoring and diagnosis system based on a cloud platform to collect and analyze the operating status of each scene power module in real time, and use machine learning algorithms to predict the trends of various performance indicators. When abnormal deviations from performance indicators or signs of failure are found, corresponding early warning information and maintenance suggestions are automatically generated and sent to relevant operation and maintenance personnel in the form of emails and text messages to guide them to perform preventive maintenance, so as to ensure the long-term stable operation of the power module.

[0153] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0154] In one embodiment, Figure 2 As shown, a standardized power module design system for different application scenarios is provided, wherein the standardized power module includes a passive filter device and a fault protection circuit; the system includes:

[0155] Standardized design module 1, used to obtain power module parameter index data of multiple target scenarios, and perform cluster analysis on all power module parameter index data to generate several benchmark module design solutions;

[0156] Lightweight design module 2 is used to perform lightweight optimization design on the passive filter in the corresponding reference module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, and generate the corresponding filter manufacturing process file;

[0157] Reliability design module 3, used to perform module fault simulation analysis on each reference module design scheme and generate corresponding fault protection circuit design files;

[0158] The optimal solution generation module 4 is used to generate the optimal standardized module design solution for the corresponding scenario based on the preset module performance indicator system according to the benchmark module design solution of each target scenario, the filter manufacturing process file and the fault protection circuit design file.

[0159] In one embodiment, the power module parameter index data includes electrical performance index, reliability index and environmental adaptability index; the power module parameter index data under all target scenarios are clustered and analyzed to generate several reference module design schemes, including:

[0160] The electrical performance indicators, reliability indicators and environmental adaptability indicators in all target scenarios are preprocessed to obtain the corresponding standardized parameter indicator data sets;

[0161] According to the standardized parameter indicator data set, key parameter indicator screening is performed based on correlation analysis and principal component analysis to obtain a corresponding key parameter indicator data set;

[0162] K-means clustering analysis is performed on the key parameter indicator data set to obtain several benchmark module design solutions.

[0163] In one embodiment, after performing cluster analysis on all power module parameter index data to generate a plurality of reference module design solutions, the method further includes:

[0164] According to each benchmark module design scheme, establish the corresponding power module simulation model;

[0165] Perform multi-physics field simulation analysis on each power module simulation model to obtain the corresponding module performance evaluation results;

[0166] According to the module performance evaluation results, the structural parameters and topological parameters of the corresponding power module simulation model are optimized to generate a module selection guidance strategy;

[0167] Abstracting each reference module design scheme based on parametric modeling technology to generate a corresponding module parametric model; the module parametric model includes structural parameters, device parameters and material parameters;

[0168] Each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model are stored in a preset relational database to generate a corresponding power module knowledge base.

[0169] In one embodiment, according to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding reference module design scheme is lightweight optimized and designed to generate the corresponding filter manufacturing process file, including:

[0170] According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structure model of the corresponding passive filter is established;

[0171] According to the initial structural model, a component material property data set, a target volume range, and a target weight range of the corresponding passive filter are obtained; the component material property data set includes material data and shape data of various components;

[0172] Performing component size prediction analysis according to the component material property data set, the target volume range, the target weight range and a pre-constructed filter optimal size prediction model to generate a corresponding filter component optimal size set; the filter optimal size prediction model is obtained based on a preset feedforward neural network training;

[0173] According to the optimal size set of the filter element, the initial structural model is optimized to generate a corresponding first optimized structural model;

[0174] By using electromagnetic field simulation software, simulation analysis of electromagnetic field distribution and filtering performance of the first optimized structural model is performed to generate corresponding simulation analysis results;

[0175] According to the simulation analysis result, fine-tuning the corresponding first optimized structural model to generate a corresponding second optimized structural model;

[0176] The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; the filter manufacturing process file includes component materials, component size specifications and processing technology.

[0177] In one embodiment, the construction of the filter optimal size prediction model includes:

[0178] Acquire a passive filter design parameter data set; the passive filter design parameter data set includes component material property data, component size data, and corresponding device volume weight data of the passive filter under different target scenarios;

[0179] The preset feedforward neural network is optimized and trained using component material property data and corresponding component size data as model input and model output respectively, and the corresponding device volume weight data as optimization constraints to obtain the filter optimal size prediction model.

[0180] In one embodiment, the step of performing module fault simulation analysis on each reference module design scheme and generating a corresponding fault protection circuit design file comprises:

[0181] According to the preset fault conditions, module fault simulation is performed on each benchmark module design scheme, and the corresponding module fault operation data set is collected;

[0182] Preprocessing the module fault operation data to generate a module status feature data set;

[0183] According to the module status feature data set, a support vector machine model is trained to construct a corresponding module fault detection model;

[0184] The fault protection circuit design file is generated according to the module fault detection model and the corresponding fault condition processing strategy.

[0185] In one embodiment, the preset module performance index system includes a standardization degree index, a lightweight index and a reliability index; the reference module design scheme according to each target scenario, the filter manufacturing process file and the fault protection circuit design file, and the corresponding scenario optimal standardized module design scheme is generated based on the preset module performance index system, including:

[0186] Taking the maximization of the module's continuous working power and energy utilization efficiency as the optimization goals, a corresponding scenario multi-objective optimization model is established based on the preset scenario voltage range and current range constraints of each target scenario;

[0187] By means of the hierarchical analysis method, the initial weight vector of the indicators corresponding to the preset module performance indicator system under each target scenario is obtained, and the fitness function is constructed according to the weighted average value of each module performance indicator in the preset module performance indicator system;

[0188] The reference module design scheme of each target scenario and the corresponding filter manufacturing process file and the fault protection circuit design file are combined and designed to generate the corresponding scenario initial module design scheme, and the module technical parameters and module performance index data corresponding to the scenario initial module design scheme are obtained; the module technical parameters include output voltage, output current and input power;

[0189] According to the module technical parameters, the module performance index data and the initial index weight vector of the initial module design scheme of each scenario, a genetic algorithm is executed based on the fitness function to solve the scenario multi-objective optimization model to obtain the corresponding scenario optimal index weight vector;

[0190] According to the optimal indicator weight vector of each scenario, the parameters of the initial module design scheme of the corresponding scenario are optimized to generate the optimal standardized module design scheme of the corresponding scenario.

[0191] In one embodiment, the system further comprises:

[0192] The adaptive optimization module is used to obtain the environmental condition data of each target scene, perform environmental adaptability simulation analysis on the optimal standardized module design scheme for the corresponding scene based on the environmental condition data, obtain the module units to be optimized, and adjust the design parameters and / or material selection of the units to be optimized in the optimal standardized module design scheme for the scene.

[0193] In one embodiment, the system further comprises:

[0194] The test optimization module is used to produce corresponding scenario power modules according to the scenario optimal standardized module design scheme of each target scenario, and to perform online performance detection on the scenario power module according to the pre-built performance anomaly prediction model, and to optimize and adjust the module production process according to the corresponding performance detection results.

[0195] For the specific limitations of the standardized power module design system for different application scenarios, please refer to the limitations of the standardized power module design method for different application scenarios above, and the corresponding technical effects can also be obtained equivalently, which will not be repeated here. Each module in the above-mentioned standardized power module design system for different application scenarios can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0196] Figure 3 FIG. 1 shows an internal structure diagram of a computer device in an embodiment, and the computer device may specifically be a terminal or a server. Figure 3 As shown, the computer device includes a processor, a memory, a network interface, a display, a camera and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a lightweight consensus method for distributed green electricity trading is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse.

[0197] It can be understood by those skilled in the art that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computing device may include more or less components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0198] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the above method are implemented when the processor executes the computer program.

[0199] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0200] The above-mentioned embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and interpreted scripting language JavaScript, etc.

[0201] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0202] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0203] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0204] In summary, the embodiments of the present invention provide a standardized power module design method, system, computer device and storage medium for different application scenarios. The standardized power module design method for different application scenarios realizes the acquisition of power module parameter index data for multiple target scenarios, performs cluster analysis on all power module parameter index data to generate several benchmark module design schemes, performs lightweight optimization design on the passive filter in the corresponding benchmark module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, generates the corresponding filter manufacturing process file, performs module fault simulation analysis on each benchmark module design scheme, generates the corresponding fault protection circuit design file, and generates the corresponding filter manufacturing process file according to the benchmark module design scheme for each target scenario. The invention discloses a technical scheme for generating a standardized module design scheme for a corresponding scenario based on a preset module performance index system, and a lightweight optimization design of a passive filter and an intelligent fault protection circuit design based on the preset module performance index system which considers multiple module characteristics such as standardization, lightweight and high reliability. The method can improve the design efficiency of the power module in different application scenarios, effectively balance the application requirements of standardization, lightweight and high reliability, and improve the environmental adaptability of the power module, which has high application value.

[0205] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0206] The above-mentioned embodiments only express several preferred implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention. It should be pointed out that once the technical personnel in this field know the basic creative concept, they can make other changes and modifications to these embodiments; that is, for the ordinary technicians in this technical field, without departing from the technical principle of the present invention, they can also make several improvements and substitutions, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention, and the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A standardized power module design method for different application scenarios, characterized in that: The standardized power module includes a passive filter device and a fault protection circuit; the method includes the following steps: Obtain power module parameter index data for multiple target scenarios, perform cluster analysis on all power module parameter index data, and generate several benchmark module design solutions; According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding benchmark module design solution is lightweight optimized and the corresponding filter manufacturing process file is generated; Perform module fault simulation analysis on each benchmark module design scheme and generate corresponding fault protection circuit design files; According to the reference module design scheme of each target scenario, the filter manufacturing process file and the fault protection circuit design file, the optimal standardized module design scheme for the corresponding scenario is generated based on the preset module performance indicator system; Among them, the step of performing lightweight optimization design on the passive filter in the corresponding reference module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario and generating the corresponding filter manufacturing process file includes: According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structure model of the corresponding passive filter is established; According to the initial structural model, a component material property data set, a target volume range, and a target weight range of the corresponding passive filter are obtained; the component material property data set includes material data and shape data of various components; Performing component size prediction analysis according to the component material property data set, the target volume range, the target weight range and a pre-constructed filter optimal size prediction model to generate a corresponding filter component optimal size set; the filter optimal size prediction model is obtained based on a preset feedforward neural network training; According to the optimal size set of the filter element, the initial structural model is optimized to generate a corresponding first optimized structural model; By using electromagnetic field simulation software, simulation analysis of electromagnetic field distribution and filtering performance of the first optimized structural model is performed to generate corresponding simulation analysis results; According to the simulation analysis result, fine-tuning the corresponding first optimized structural model to generate a corresponding second optimized structural model; The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; the filter manufacturing process file includes component materials, component size specifications and processing technology.

2. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The power module parameter index data includes electrical performance index, reliability index and environmental adaptability index; The steps of clustering the power module parameter index data under all target scenarios and generating several benchmark module design solutions include: The electrical performance indicators, reliability indicators and environmental adaptability indicators in all target scenarios are preprocessed to obtain the corresponding standardized parameter indicator data sets; According to the standardized parameter indicator data set, key parameter indicators are screened based on correlation analysis and principal component analysis to obtain a corresponding key parameter indicator data set; K-means clustering analysis is performed on the key parameter indicator data set to obtain several benchmark module design solutions.

3. The standardized power module design method for different application scenarios as claimed in claim 2, characterized in that: After the step of clustering and analyzing all power module parameter index data to generate a plurality of reference module design solutions, the method further includes: According to each benchmark module design scheme, establish the corresponding power module simulation model; Perform multi-physics field simulation analysis on each power module simulation model to obtain the corresponding module performance evaluation results; According to the module performance evaluation results, the structural parameters and topological parameters of the corresponding power module simulation model are optimized to generate a module selection guidance strategy; Abstracting each reference module design scheme based on parametric modeling technology to generate a corresponding module parametric model; the module parametric model includes structural parameters, device parameters and material parameters; Each benchmark module design scheme and the corresponding module selection guidance strategy and module parameterization model are stored in a preset relational database to generate a corresponding power module knowledge base.

4. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The steps of constructing the filter optimal size prediction model include: Acquire a passive filter design parameter data set; the passive filter design parameter data set includes component material property data, component size data, and corresponding device volume weight data of the passive filter under different target scenarios; The preset feedforward neural network is optimized and trained using component material property data and corresponding component size data as model input and model output respectively, and the corresponding device volume weight data as optimization constraints to obtain the filter optimal size prediction model.

5. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The step of performing module fault simulation analysis on each reference module design scheme to generate a corresponding fault protection circuit design file comprises: According to the preset fault conditions, module fault simulation is performed on each benchmark module design scheme, and the corresponding module fault operation data set is collected; Preprocessing the module fault operation data to generate a module status feature data set; According to the module status feature data set, a support vector machine model is trained to construct a corresponding module fault detection model; The fault protection circuit design file is generated according to the module fault detection model and the corresponding fault condition processing strategy.

6. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The preset module performance index system includes a standardization index, a lightweight index and a reliability index; The step of generating the optimal standardized module design scheme for the corresponding scenario based on the preset module performance indicator system according to the reference module design scheme for each target scenario, the filter manufacturing process file and the fault protection circuit design file comprises: Taking the maximization of the module's continuous working power and energy utilization efficiency as the optimization goals, a corresponding scenario multi-objective optimization model is established based on the preset scenario voltage range and current range constraints of each target scenario; By means of the hierarchical analysis method, the initial weight vector of the indicators corresponding to the preset module performance indicator system under each target scenario is obtained, and the fitness function is constructed according to the weighted average value of each module performance indicator in the preset module performance indicator system; The reference module design scheme of each target scenario and the corresponding filter manufacturing process file and the fault protection circuit design file are combined and designed to generate the corresponding scenario initial module design scheme, and the module technical parameters and module performance index data corresponding to the scenario initial module design scheme are obtained; the module technical parameters include output voltage, output current and input power; According to the module technical parameters, the module performance index data and the initial index weight vector of the initial module design scheme of each scenario, a genetic algorithm is executed based on the fitness function to solve the scenario multi-objective optimization model to obtain the corresponding scenario optimal index weight vector; According to the optimal indicator weight vector of each scenario, the parameters of the initial module design scheme of the corresponding scenario are optimized to generate the optimal standardized module design scheme of the corresponding scenario.

7. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The method further comprises: Acquire environmental condition data for each target scenario, perform environmental adaptability simulation analysis on the optimal standardized module design scheme for the corresponding scenario based on the environmental condition data, obtain the module units to be optimized, and adjust the design parameters and / or material selection of the units to be optimized in the optimal standardized module design scheme for the scenario.

8. The standardized power module design method for different application scenarios as claimed in claim 1, characterized in that: The method further comprises: According to the scene-optimal standardized module design plan for each target scene, the corresponding scene power module is produced, and the scene power module is subjected to online performance testing according to the pre-built performance anomaly prediction model, and the module production process is optimized and adjusted according to the corresponding performance test results.

9. A standardized power module design system for different application scenarios, characterized in that: The standardized power module includes a passive filter device and a fault protection circuit; the system includes: Standardized design module, used to obtain power module parameter index data for multiple target scenarios, and perform cluster analysis on all power module parameter index data to generate several benchmark module design solutions; The lightweight design module is used to optimize the passive filter in the corresponding benchmark module design scheme according to the module electrical parameter requirements and filtering performance requirements of each target scenario, and generate the corresponding filter manufacturing process file; Reliability design module, used to perform module fault simulation analysis on each reference module design scheme and generate corresponding fault protection circuit design files; An optimal solution generation module, used to generate an optimal standardized module design solution for the corresponding scenario based on a preset module performance indicator system according to the reference module design solution of each target scenario, the filter manufacturing process file and the fault protection circuit design file; Among them, according to the module electrical parameter requirements and filtering performance requirements of each target scenario, the passive filter in the corresponding benchmark module design scheme is lightweight optimized and designed to generate the corresponding filter manufacturing process file, including: According to the module electrical parameter requirements and filtering performance requirements of each target scenario, the initial structure model of the corresponding passive filter is established; According to the initial structural model, a component material property data set, a target volume range, and a target weight range of the corresponding passive filter are obtained; the component material property data set includes material data and shape data of various components; Performing component size prediction analysis according to the component material property data set, the target volume range, the target weight range and a pre-constructed filter optimal size prediction model to generate a corresponding filter component optimal size set; the filter optimal size prediction model is obtained based on a preset feedforward neural network training; According to the optimal size set of the filter element, the initial structural model is optimized to generate a corresponding first optimized structural model; By using electromagnetic field simulation software, simulation analysis of electromagnetic field distribution and filtering performance of the first optimized structural model is performed to generate corresponding simulation analysis results; According to the simulation analysis result, fine-tuning the corresponding first optimized structural model to generate a corresponding second optimized structural model; The filter manufacturing process file is generated according to the structural parameters of the second optimized structural model; the filter manufacturing process file includes component materials, component size specifications and processing technology.

Images