Dry-type transformer structure parameter optimization method and device, equipment and storage medium

Abstract

The present application relates to the technical field of dry-type transformer structure parameter optimization, and particularly relates to a dry-type transformer structure parameter optimization method, device, equipment and storage medium, the method first acquires a first initial structure parameter set, performs a simulation operation based on the first initial structure parameter set to obtain a first to-be-optimized parameter set, then performs parameter optimization processing on the first to-be-optimized parameter set to obtain a first optimized structure parameter set, then acquires a second to-be-optimized parameter set and a preset optimization target, performs multi-round iteration optimization processing on the second to-be-optimized parameter set based on the optimization target by using a multi-objective genetic algorithm to obtain a second optimized structure parameter set, and finally integrates the first optimized structure parameter set and the second optimized structure parameter set to obtain a target optimized structure parameter set, which realizes precise control of product volume and weight while strengthening the short-circuit mechanical strength of the transformer, improving the anti-deformation capability of the winding and the core, and achieving synchronous optimization of low-loss and low-noise indicators.

Description

Technical Field

[0001] This invention relates to the field of dry-type transformer structural parameter optimization technology, and in particular to a method, apparatus, equipment and storage medium for optimizing dry-type transformer structural parameters. Background Technology

[0002] In industrial production and power transmission, the reliability and adaptability of dry-type transformers directly affect the stable operation of the system. With the diversification of application scenarios, their design faces multiple core contradictions that urgently need to be addressed. On the one hand, short-circuit faults in power systems occur frequently, making it crucial to strengthen the transformer's mechanical strength against short circuits to ensure operational safety. This requires optimizing the structure of windings, cores, and support frames to improve deformation resistance. However, traditional structural optimization often results in increased size and weight, conflicting with the installation requirements for miniaturization and compactness. How to accurately control size and weight while enhancing mechanical strength has become a significant bottleneck restricting product adaptability. On the other hand, energy saving and noise reduction have become the core trend in transformer design. The simultaneous achievement of low loss and low noise indicators faces significant constraints. Although magnetic circuit optimization can reduce iron loss and copper loss, it may exacerbate magnetostrictive vibration of the iron core, thereby increasing the noise level. Circuit parameter adjustment and structural damping design, while suppressing noise, may also affect magnetic circuit energy efficiency and loss control. The contradiction between magnetic circuit, circuit and structural optimization makes it difficult to achieve low loss and low noise in a coordinated manner. Existing design solutions often fail to meet multi-dimensional requirements and fail to achieve a balance between mechanical performance, compactness requirements and energy saving and environmental protection indicators. Summary of the Invention

[0003] In order to overcome the shortcomings of the prior art, the present invention aims to provide a method, device, equipment and storage medium for optimizing the structural parameters of dry-type transformers, which realizes the simultaneous optimization of product volume and weight while strengthening the transformer's short-circuit mechanical strength and improving the winding and core's resistance to deformation, and achieving low loss and low noise indicators.

[0004] The first aspect of the present invention provides a method for optimizing the structural parameters of a dry-type transformer, comprising: obtaining a first initial set of structural parameters; performing a simulation operation based on the first initial set of structural parameters to obtain a first set of parameters to be optimized; performing parameter optimization processing on the first set of parameters to be optimized to obtain a first optimized set of structural parameters; obtaining a second set of parameters to be optimized and a preset optimization objective; using a multi-objective genetic algorithm to perform multiple rounds of iterative optimization processing on the second set of parameters to be optimized based on the optimization objective to obtain a second optimized set of structural parameters; and integrating the first optimized set of structural parameters and the second optimized set of structural parameters to obtain a target optimized set of structural parameters.

[0005] Optionally, in a first implementation of the first aspect of the present invention, the first initial structural parameter set includes winding copper flat wire cross-sectional parameters, core lamination thickness parameters, and winding preload parameters; the step of performing simulation operations based on the first initial structural parameter set to obtain a first set of parameters to be optimized includes: constructing an initial three-dimensional simulation model using simulation tools; inputting the winding copper flat wire cross-sectional parameters and the core lamination thickness parameters into the initial three-dimensional simulation model for simulation calculation processing to obtain initial electromagnetic force distribution parameters; inputting the initial electromagnetic force distribution parameters and the winding preload parameters into the initial three-dimensional simulation model for simulation calculation processing to obtain initial winding deformation parameters and initial core displacement parameters; and integrating the initial electromagnetic force distribution parameters, the initial winding deformation parameters, and the initial core displacement parameters to obtain the first set of parameters to be optimized.

[0006] Optionally, in a second implementation of the first aspect of the present invention, the step of performing parameter optimization processing on the first set of parameters to be optimized to obtain a first set of optimized structural parameters includes: determining the structural weak points of the initial three-dimensional simulation model based on the initial electromagnetic force distribution parameters; determining the short-circuit withstand strength failure items of the initial three-dimensional simulation model based on the initial winding deformation parameters and the initial core displacement parameters; performing parameter optimization processing on the initial three-dimensional simulation model based on the structural weak points and the short-circuit withstand strength failure items to obtain an optimized three-dimensional simulation model; inputting the winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters into the optimized three-dimensional simulation model for simulation calculation processing to obtain optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters; and integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first set of optimized structural parameters.

[0007] Optionally, in a third implementation of the first aspect of the present invention, the step of integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first optimized structural parameter set includes: obtaining a preset winding deformation threshold and a preset volumetric weight threshold; comparing the optimized winding deformation parameters with the winding deformation threshold; obtaining volumetric weight parameters based on the optimized three-dimensional simulation model, and comparing the volumetric weight parameters with the volumetric weight threshold; when the optimized winding deformation parameters are less than or equal to the winding deformation threshold, and the volumetric weight parameters are less than or equal to the volumetric weight threshold, integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first optimized structural parameter set.

[0008] Optionally, in a fourth implementation of the first aspect of the present invention, the step of using a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain a second optimized structural parameter set includes: using the multi-objective genetic algorithm to randomly generate multiple sets of initial candidate parameters based on the second set of parameters to be optimized; calling a pre-trained coupled model to perform performance evaluation on the multiple sets of initial candidate parameters respectively to obtain a performance evaluation parameter set corresponding to each set of initial candidate parameters; performing screening processing on the multiple sets of initial candidate parameters based on the optimization objective and the multiple sets of performance evaluation parameter sets to obtain multiple sets of high-quality candidate parameters; using the multi-objective genetic algorithm to perform multiple rounds of selection, crossover, and mutation processing on the multiple sets of high-quality candidate parameters to obtain a high-quality candidate parameter set that satisfies the optimization objective, and using the high-quality candidate parameter set as the second optimized structural parameter set.

[0009] A second aspect of the present invention provides a device for optimizing the structural parameters of a dry-type transformer, comprising: a simulation module for acquiring a first initial set of structural parameters and performing simulation operations based on the first initial set of structural parameters to obtain a first set of parameters to be optimized; a first optimization module for performing parameter optimization processing on the first set of parameters to be optimized to obtain a first optimized set of structural parameters; a second optimization module for acquiring a second set of parameters to be optimized and a preset optimization objective, and using a multi-objective genetic algorithm to perform multiple rounds of iterative optimization processing on the second set of parameters to be optimized based on the optimization objective to obtain a second optimized set of structural parameters; and a parameter integration module for integrating the first optimized set of structural parameters and the second optimized set of structural parameters to obtain a target optimized set of structural parameters.

[0010] A third aspect of the present invention provides a device for optimizing the structural parameters of a dry-type transformer, the device comprising: a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to cause the device to execute the steps of the dry-type transformer structural parameter optimization method described in any of the preceding claims.

[0011] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the dry-type transformer structural parameter optimization method described in any of the preceding claims.

[0012] In the technical solution of this invention, a first initial structural parameter set is first obtained, and a simulation operation is performed based on the first initial structural parameter set to obtain a first parameter set to be optimized. Then, the first parameter set to be optimized is optimized to obtain a first optimized structural parameter set. Next, a second parameter set to be optimized and a preset optimization target are obtained. A multi-objective genetic algorithm is used to perform multiple rounds of iterative optimization on the second parameter set to be optimized based on the optimization target to obtain a second optimized structural parameter set. Finally, the first optimized structural parameter set and the second optimized structural parameter set are integrated to obtain the target optimized structural parameter set. This achieves the simultaneous optimization of low loss and low noise indicators while strengthening the transformer's short-circuit mechanical strength and improving the winding and core's resistance to deformation. Attached Figure Description

[0013] Figure 1 A logic flowchart of the dry-type transformer structural parameter optimization method provided in an embodiment of the present invention;

[0014] Figure 2 This is a schematic diagram of the structure of the dry-type transformer structural parameter optimization device provided in an embodiment of the present invention;

[0015] Figure 3 A schematic diagram of the structure of the dry-type transformer structural parameter optimization device provided in an embodiment of the present invention. Detailed Implementation

[0016] This invention provides a method, apparatus, device, and storage medium for optimizing the structural parameters of a dry-type transformer. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the method for optimizing the structural parameters of a dry-type transformer in this invention includes:

[0018] 101. Obtain the first initial structural parameter set, and perform simulation operations based on the first initial structural parameter set to obtain the first parameter set to be optimized;

[0019] In this embodiment, the core of the first initial structural parameter set covers the key structural design parameters of the dry-type transformer, specifically including the winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters. This parameter set is the core fundamental parameter affecting the mechanical characteristics and electromagnetic response of the transformer's initial structure. The simulation operation was conducted using the Ansys Maxwell and Mechanical co-simulation platform. First, a three-dimensional finite element model was constructed to ensure a high degree of consistency between the model and the geometric features and mechanical properties of the actual structure. Then, the first initial structural parameters were imported into the model, and multi-dimensional simulation analysis was performed under short-circuit current conditions to obtain the first set of parameters to be optimized. Through the co-simulation platform and integrated three-dimensional modeling technology, the accurate characterization of the mechanical performance of the initial structure under short-circuit conditions was achieved, effectively avoiding the subjectivity and bias of structural performance prediction in traditional empirical design, and providing an objective quantitative basis for determining the subsequent optimization direction.

[0020] 102. Perform parameter optimization processing on the first set of parameters to be optimized to obtain the first optimized structural parameter set;

[0021] In this embodiment, the first set of parameters to be optimized is processed to obtain the first set of optimized structural parameters. The core is to accurately locate the shortcomings in the short-circuit withstand performance and the weak points in the structure of the dry-type transformer based on the electromagnetic force distribution data, winding deformation data, and core displacement data in the set of parameters to be optimized, and then carry out targeted parameter optimization. For example, firstly, key performance data is extracted from the first set of parameters to be optimized to identify the core problems in the initial three-dimensional simulation model structure, such as the stress concentration area caused by the 90° splicing of the core and the excessive winding deformation. The optimization direction is determined with the adjustment of the core splicing parameters as the core. Based on this optimization direction, the initial 90° splicing parameters of the core laminations are adjusted to 45° oblique splicing parameters. The core logic is that the 45° oblique splicing can change the force transmission path between the laminations, disperse the concentrated stress at the splicing point, effectively alleviate the local overload problem caused by the 90° splicing, thereby reducing the winding deformation, and finally forming an optimized three-dimensional simulation model. The first set of optimized structural parameters is obtained based on the optimized three-dimensional simulation model. The quantitative data based on the parameter set to be optimized accurately pinpoints the root cause of the problem, making the optimization direction more targeted. This significantly improves the transformer's short-circuit mechanical strength and structural stability, ensuring operational reliability under extreme conditions. Compared with traditional empirical optimization methods, this process relies on quantitative data and simulation verification, significantly improving optimization efficiency and reducing the cost of physical prototype trial and error, providing a scientific and feasible technical path for transformer structure optimization.

[0022] 103. Obtain the second set of parameters to be optimized and the preset optimization objective, and use a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain the second optimized structure parameter set;

[0023] In this embodiment, the second set of parameters to be optimized covers the core design parameters of the transformer's magnetic circuit and electrical circuit, mainly including key parameters such as peak magnetic flux density, coil resistance, and magnetic circuit air gap. It also incorporates relevant adaptation parameters related to the segmented design of the magnetic circuit and the coil damping structure, forming a basic parameter system supporting the coordinated optimization of low loss and low noise. The predetermined optimization targets are determined based on national and industry technical specifications for high-efficiency energy-saving power equipment, combined with the energy-saving and environmental protection requirements of the transformer's rated operating conditions and application scenarios. Furthermore, through technical feasibility analysis and performance requirement surveys, the no-load loss threshold, load loss threshold, and noise threshold are clarified. For example, the no-load loss is set to not exceed 1.035kW, the load loss not to exceed 6.715kW, and the noise not to exceed 55dB. The optimization targets are scientifically determined based on industry standards and actual needs, making the parameter optimization direction more targeted and ensuring that the optimization results can effectively meet the dual requirements of engineering applications and energy conservation and environmental protection. The second set of parameters to be optimized covers the core design parameters of the transformer's magnetic circuit and electrical circuit, mainly including key parameters such as peak magnetic flux density, coil resistance, and air gap in the magnetic circuit. It also incorporates relevant adaptation parameters related to the segmented design of the magnetic circuit and the coil damping structure, forming a basic parameter system supporting the coordinated optimization of low loss and low noise. During the optimization process, firstly, based on the second set of parameters to be optimized, multiple initial candidate parameter sets are randomly generated using the NSGA-II multi-objective genetic algorithm. Then, a pre-trained coupled model is called, and each candidate parameter set is imported into the model for performance quantification evaluation. The model outputs loss values, noise sound pressure levels, and magnetic circuit energy efficiency ratio data for the corresponding parameter combinations, providing an objective basis for parameter selection. Based on the preset optimization objectives, the evaluation results are screened, retaining high-quality candidate parameter sets that meet the performance standards. Then, through non-dominated sorting and crowding calculation using the multi-objective genetic algorithm, parameter combinations that combine performance advantages and diversity are selected for the iteration phase. Through genetic operations such as selection, crossover, and mutation, a new generation of candidate parameter sets is generated. The coupled model evaluation and objective selection are repeated, gradually approaching the optimal solution through multiple iterations until a parameter combination that satisfies all optimization objectives is obtained. Finally, these are integrated to form the second optimized structural parameter set. By leveraging the global search capability of the multi-objective genetic algorithm and the precise evaluation characteristics of the coupled model, the constraint relationship between low loss and low noise was effectively resolved. Through multiple rounds of iteration and quantitative evaluation, performance imbalance caused by single parameter optimization was avoided, and the scientificity and reliability of parameter optimization were improved. In addition, the synergistic application of the algorithm and model replaced the traditional empirical trial-and-error design, which significantly shortened the R&D cycle and reduced trial-and-error costs. The optimized second set of structural parameters provides a precise basis for the engineering realization of the low loss and low noise characteristics of the transformer.

[0024] 104. Integrate the first optimized structural parameter set and the second optimized structural parameter set to obtain the target optimized structural parameter set.

[0025] In this embodiment, on the one hand, the target optimized structural parameter set fully retains the strengthening effect of the first optimized structural parameter set on the transformer's short-circuit mechanical strength, significantly improving the winding and core's resistance to deformation. This achieves precise control of product volume and weight while enhancing mechanical performance, fully meeting the requirements for compact equipment installation and resolving the contradiction between traditional mechanical strength enhancement design and volume and weight control. On the other hand, the target optimized structural parameter set integrates the magnetic and electrical circuit optimization results of the second optimized structural parameter set. While ensuring mechanical performance and compactness requirements, it achieves simultaneous achievement of no-load loss, load loss, and noise indicators, effectively balancing the inherent constraints of magnetic circuit, electrical circuit, and structural optimization. The target optimized structural parameter set constructs a parameter benchmark for all-dimensional performance assurance, avoiding performance imbalances caused by single-dimensional optimization. This allows for simultaneous improvement in the transformer's operational reliability, compact adaptability, and energy-saving and environmental protection characteristics under extreme operating conditions, significantly overcoming the technical bottleneck of balancing multiple objectives in traditional design. It provides accurate and feasible parameter basis for the engineering realization of high-performance dry-type transformers.

[0026] In this embodiment of the invention, the first initial structural parameter set includes winding copper flat wire cross-sectional parameters, core lamination thickness parameters, and winding preload parameters. The step of performing simulation operations based on the first initial structural parameter set to obtain a first set of parameters to be optimized includes: constructing an initial three-dimensional simulation model using simulation tools; inputting the winding copper flat wire cross-sectional parameters and the core lamination thickness parameters into the initial three-dimensional simulation model for simulation calculation to obtain initial electromagnetic force distribution parameters; inputting the initial electromagnetic force distribution parameters and the winding preload parameters into the initial three-dimensional simulation model for simulation calculation to obtain initial winding deformation parameters and initial core displacement parameters; and integrating the initial electromagnetic force distribution parameters, the initial winding deformation parameters, and the initial core displacement parameters to obtain the first set of parameters to be optimized.

[0027] In this embodiment, the winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters covered by the first initial structural parameter set are the core fundamental parameters that determine the initial electromagnetic characteristics and mechanical properties of the dry-type transformer. First, an initial three-dimensional simulation model that closely matches the geometric features and material properties of the actual product needs to be constructed using simulation tools. This initial three-dimensional simulation model must completely replicate the spatial structural relationships of the winding, core, and support frame to ensure the reliability and validity of the simulation results. The simulation tool can be the Ansys Maxwell and Mechanical co-simulation platform. Subsequently, simulation calculations were carried out in stages. In the first stage, the cross-sectional parameters of the copper flat wire of the winding and the thickness parameters of the laminations of the iron core were input into the model. Based on the principle of electromagnetic induction and the finite element analysis method, the electromagnetic interaction between the winding and the iron core under the action of short-circuit current was simulated to obtain the initial electromagnetic force distribution parameters. These parameters directly reflect the stress intensity and distribution law of each part under short-circuit conditions. In the second stage, based on the obtained initial electromagnetic force distribution parameters, the winding preload parameters were further input to carry out structural mechanics simulation calculations. The deformation degree of the winding and the displacement of the iron core under the combined action of electromagnetic force and preload were analyzed to obtain the initial winding deformation parameters and the initial iron core displacement parameters, thereby quantifying the deformation resistance of the initial structure. Finally, by integrating the above initial electromagnetic force distribution parameters, initial winding deformation parameters, and initial iron core displacement parameters, a first set of parameters to be optimized was formed that comprehensively reflects the electromagnetic and mechanical coupling performance of the initial structure under short-circuit conditions.

[0028] In this embodiment of the invention, the step of performing parameter optimization processing on the first set of parameters to be optimized to obtain a first set of optimized structural parameters includes: determining the structural weak points of the initial three-dimensional simulation model based on the initial electromagnetic force distribution parameters; determining the short-circuit withstand strength failure items of the initial three-dimensional simulation model based on the initial winding deformation parameters and the initial core displacement parameters; performing parameter optimization processing on the initial three-dimensional simulation model based on the structural weak points and the short-circuit withstand strength failure items to obtain an optimized three-dimensional simulation model; inputting the winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters into the optimized three-dimensional simulation model for simulation calculation processing to obtain optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters; and integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first set of optimized structural parameters.

[0029] In this embodiment, an in-depth analysis is first performed based on the initial electromagnetic force distribution parameters. Combined with structural mechanics principles, regions with concentrated electromagnetic forces are identified. These regions are prone to structural failure due to overload. Based on this, the weak points in the initial 3D simulation model are determined, such as core splices and winding ends—areas with large electromagnetic force gradients. Subsequently, the initial winding deformation parameters and initial core displacement parameters are compared with preset short-circuit withstand strength design limits. Performance indicators exceeding these limits are identified, clarifying the short-circuit withstand strength deficiencies of the initial 3D simulation model, such as excessive maximum winding deformation and excessive core displacement. Based on the identified structural weaknesses and short-circuit withstand inadequacies, targeted parameter optimization strategies were developed. By adjusting key structural parameters, such as optimizing the core splicing method and adapting the winding preload value, the initial 3D simulation model was reconstructed and optimized to obtain an optimized 3D simulation model. This ensured that optimization measures precisely targeted performance shortcomings. In-depth analysis of initial parameters enabled accurate location of structural weaknesses and short-circuit withstand inadequacies, giving subsequent optimization measures a clear focus and effectively avoiding the drawbacks of blindly adjusting parameters in traditional optimization, thus improving optimization efficiency and effectiveness. To fully verify the optimization effect, the winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters were individually input into the optimized 3D simulation model. Electromagnetic and mechanical phased simulation calculations were conducted to sequentially obtain the optimized electromagnetic force distribution parameters, winding deformation parameters, and core displacement parameters, ensuring that the performance corresponding to each core parameter was effectively improved. Finally, by integrating the above-mentioned optimized parameters, a first optimized structural parameter set is formed that can accurately reflect the mechanical performance of the optimized structure, providing a quantitative basis for the structural design of actual products. Based on the problem-oriented parameter optimization strategy and simulation model reconstruction, the shortcomings of the mechanical performance of the initial structure can be solved from the root, significantly enhancing the transformer's short-circuit resistance and improving the deformation resistance of the windings and core. The entire process relies on the simulation model to complete parameter optimization and effect verification, eliminating the need to build physical prototypes for repeated experiments, greatly reducing R&D trial and error costs and shortening the product development cycle.

[0030] In this embodiment of the invention, the step of integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first optimized structural parameter set includes: obtaining a preset winding deformation threshold and a preset volumetric weight threshold; comparing the optimized winding deformation parameters with the winding deformation threshold; obtaining volumetric weight parameters based on the optimized three-dimensional simulation model, and comparing the volumetric weight parameters with the volumetric weight threshold; when the optimized winding deformation parameters are less than or equal to the winding deformation threshold, and the volumetric weight parameters are less than or equal to the volumetric weight threshold, integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters to obtain the first optimized structural parameter set.

[0031] In this embodiment, the preset winding deformation threshold and volumetric weight threshold are quantitative benchmarks determined after technical feasibility studies and performance index breakdown, based on national power equipment industry standards, transformer rated operating condition requirements, and the compactness requirements of actual installation scenarios. The winding deformation threshold directly corresponds to the core design requirement of short-circuit withstand strength, while the volumetric weight threshold matches the installation adaptation requirements for equipment miniaturization. After completing the construction of the optimized 3D simulation model and the simulation calculation of optimized parameters, the optimized winding deformation parameters are first quantitatively compared with the preset winding deformation threshold to verify whether the optimized winding's deformation resistance meets the standards. Simultaneously, based on the geometric features and material properties of the optimized 3D simulation model, the corresponding volumetric weight parameters are extracted and calculated, and compared with the preset volumetric weight threshold to verify whether the optimized structure meets the compactness control requirements. Only when both comparison results meet the preset conditions—that is, the optimized winding deformation parameter is less than or equal to the winding deformation threshold and the volumetric weight parameter is less than or equal to the volumetric weight threshold—is it considered that the optimized structure has both improved short-circuit resistance mechanical strength and precisely controlled volume and weight. At this point, the optimized electromagnetic force distribution parameter, optimized winding deformation parameter, and optimized core displacement parameter are integrated to form a first optimized structural parameter set with practical engineering application value. Dual-dimensional comparison verification based on preset quantitative thresholds provides a scientific and unified standard for evaluating the optimization effect, effectively avoiding the subjective bias of empirical judgment and ensuring the performance reliability of the first optimized structural parameter set. Furthermore, by simultaneously verifying short-circuit resistance strength and volumetric weight indicators, a synergistic balance between mechanical performance enhancement and compactness requirements is achieved, resolving the inherent contradiction between improved mechanical strength and increased volume in traditional optimization. The first optimized structural parameter set, formed by integrating the data after meeting the standards, provides a precise quantitative basis for the structural processing and assembly of actual products, ensuring that the optimization effect can be effectively implemented.

[0032] In this embodiment of the invention, the step of using a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain a second optimized structural parameter set includes: randomly generating multiple sets of initial candidate parameter sets based on the second set of parameters to be optimized using the multi-objective genetic algorithm; calling a pre-trained coupled model to evaluate the performance of the multiple sets of initial candidate parameter sets respectively, obtaining a performance evaluation parameter set corresponding to each set of initial candidate parameter sets; screening the multiple sets of initial candidate parameter sets based on the optimization objective and the multiple sets of performance evaluation parameter sets to obtain multiple sets of high-quality candidate parameter sets; and performing multiple rounds of selection, crossover, and mutation processing on the multiple sets of high-quality candidate parameter sets using the multi-objective genetic algorithm to obtain a high-quality candidate parameter set that satisfies the optimization objective, and using the high-quality candidate parameter set as the second optimized structural parameter set.

[0033] In this embodiment, the second set of parameters to be optimized is constructed around the optimization requirements. Its core encompasses the key input parameters required by the coupled model, including core electromagnetic and structural parameters such as peak magnetic flux density, coil resistance, and magnetic circuit air gap. These parameters are directly related to loss and noise performance, forming the basic parameter system for multi-objective optimization. The pre-trained coupled model adopts a collaborative architecture of an electromagnetic calculation module, a loss calculation module, a noise calculation module, and a data fusion module. This architecture can integrate and characterize the correlation characteristics of multiple physical fields. The electromagnetic calculation module is the fundamental core of the entire coupled model. Based on the input core parameters such as peak magnetic flux density, magnetic circuit air gap, and coil resistance, and combined with electromagnetic induction and electromagnetic field theory, it solves for the magnetic field strength distribution, electromagnetic force amplitude variation, and magnetic flux transfer efficiency inside the transformer. The output magnetic field characteristic data and electromagnetic transfer efficiency data are not only the core input sources for the loss calculation module and the noise calculation module, but also the basis for subsequent magnetic circuit energy efficiency ratio calculations. The loss calculation module uses the alternating magnetic field data and coil current density data output by the electromagnetic calculation module as its basis, relying on the principle of separating iron loss and copper loss calculation. The iron loss of the core material and the copper loss of the coil conductor are quantified separately, and the specific values ​​of no-load loss and load loss are finally integrated and output. Simultaneously, the magnetic flux transfer efficiency data provided by the electromagnetic calculation module is used to calculate the magnetic circuit energy efficiency ratio, which characterizes the energy conversion performance of the magnetic circuit. The noise calculation module analyzes the magnetostrictive vibration characteristics of the core based on the magnetic flux density alternating data output by the electromagnetic calculation module, analyzes the structural vibration response of the coil and core combined with electromagnetic force data, and then directly outputs the noise sound pressure level under the corresponding operating condition through vibration and noise transfer functions. The data fusion module establishes a correlation verification mechanism among loss values, noise sound pressure level, and magnetic circuit energy efficiency ratio to eliminate errors caused by theoretical assumptions or calculation deviations in a single module, ensuring that the three core performance parameters output by the model have high consistency and reliability. The training of the coupled model adopts a phased strategy: first, the dataset is constructed; then, each module is pre-trained independently; and finally, joint fine-tuning is carried out, using physical experiment and simulation data as training samples, and minimizing prediction errors with the help of the Adam optimizer. The predetermined optimization targets are determined based on national and industry technical specifications for high-efficiency and energy-saving power equipment, combined with the energy-saving and environmental protection requirements of transformer rated operating conditions and application scenarios. Furthermore, through technical feasibility analysis and performance requirement surveys, the no-load loss threshold, load loss threshold, and noise threshold are clearly defined. For example, the no-load loss is set to not exceed 1.035kW, the load loss not to exceed 6.715kW, and the noise not to exceed 55dB. These optimization targets are scientifically determined based on industry standards and actual needs, making parameter optimization more targeted and ensuring that the optimization results effectively meet the dual requirements of engineering applications and energy conservation and environmental protection.The optimization process follows a collaborative iterative logic of algorithm exploration and model verification: First, based on the second set of parameters to be optimized, multiple initial candidate parameter sets are constructed using the random generation mechanism of the NSGA-II multi-objective genetic algorithm. This ensures that the parameter combinations can comprehensively cover the reasonable value range of the core parameters, providing ample exploration space for discovering the global optimal solution. Then, a pre-trained coupling model is invoked, importing each initial candidate parameter set into the model for multi-physics coupling characteristic analysis. This model integrates the intrinsic correlation between electromagnetic induction, energy loss, and vibration noise, accurately outputting the loss value, noise sound pressure level, and magnetic circuit energy efficiency ratio under the corresponding parameter combinations. This forms a performance evaluation parameter set corresponding to each candidate parameter set, providing an objective basis for the quantitative verification of the optimization objective. Next, based on the preset optimization objective, the performance evaluation parameter sets corresponding to each initial candidate parameter set are systematically screened, eliminating parameter combinations that do not meet performance standards and retaining parameter schemes that can balance loss and noise requirements, resulting in multiple high-quality candidate parameter sets. Finally, the NSGA-II multi-objective genetic algorithm is used to refine the high-quality candidate parameter sets. Multiple rounds of selection, crossover, and mutation processing are conducted. Selection operations retain the best-performing parameter combinations, crossover operations fuse advantageous parameters from different superior solutions, and mutation operations appropriately expand the parameter search space to avoid local optima. After each round of crossover and mutation, the pre-trained coupled model is called again to re-evaluate the performance of the newly generated parameter combinations, outputting corresponding loss values, noise sound pressure levels, and other data to verify whether the fused parameters still meet the optimization objectives or achieve performance improvement. Based on this, a new set of high-quality candidate parameters is selected, ensuring that each iteration progresses along the direction of performance optimization, effectively overcoming the limitations of single-parameter optimization. After multiple rounds of closed-loop iteration, a set of high-quality candidate parameters that fully satisfies all optimization objectives is finally obtained and determined as the second optimized structural parameter set. This parameter set clarifies the optimal matching relationship between the core parameters of the magnetic circuit and the circuit. The entire optimization process relies on the synergy of algorithms and models to replace traditional empirical trial-and-error design, significantly shortening the R&D cycle and reducing the trial production cost of physical prototypes. Simultaneously, it provides accurate parameter basis for the engineering realization of low-loss and low-noise characteristics of transformers.

[0034] It should also be noted that the optimized matching relationship between the core parameters of the magnetic circuit and the circuit, as clearly defined in the second optimized structural parameter set, ultimately needs to be implemented through targeted physical structural design of the magnetic circuit and coils to achieve a substantial synergy between low loss and low noise. Specifically, the magnetic circuit adopts a segmented design strategy. Based on the functional requirements and performance impact of different regions of the magnetic circuit, the iron core is divided into three functional zones: the main magnetic section, the adjustment section, and the buffer section. Each zone is precisely matched and optimized through differentiated material selection and structural design. The main magnetic section uses high-permeability silicon steel, which reduces energy loss during magnetic field propagation due to the material's excellent magnetic properties. This directly addresses the iron loss problem in no-load loss and works synergistically with the optimized magnetic flux density peak value in the second optimized structural parameter set to further improve magnetic circuit efficiency. The adjustment section adds a 0.1 mm air gap, which suppresses magnetic saturation during core operation through reasonable magnetic circuit gap design. This reduces vibration and noise caused by magnetic saturation at the source and precisely matches the noise threshold control requirements. The air gap size parameter directly reflects the air gap optimization results of the second optimized structural parameter set. The buffer section uses soft magnetic composite material, which absorbs magnetostrictive vibrations to attenuate the small stretching deformation of the core caused by the magnetic field. This blocks the transmission path of vibration to the coil and shell, further enhancing the noise suppression effect and complementing the optimized magnetic circuit parameters. The coil damping design revolves around vibration attenuation and structural stability. A 2mm butyl rubber damping layer is filled between the coil and the iron core. The excellent damping properties of butyl rubber absorb the vibration energy between the coil and the iron core, reducing vibration transmission efficiency. Simultaneously, resin-impregnated fiberglass cloth is used for binding at a tension of 80 Newtons per strand. This ensures a tight fit between the coil, damping layer, and iron core, preventing frictional noise caused by relative vibration between components. It also enhances the rigidity and stability of the coil structure. Combined with the damping layer, this achieves a vibration attenuation rate of no less than 30%. The core parameters of this coil damping design, including damping layer thickness and binding tension, are derived from the optimization results of the second optimized structural parameter set. Through the dual effects of rigid constraint of the physical structure and damping absorption, the noise performance is further guaranteed to remain stable and meet standards. The aforementioned magnetic circuit segmentation and coil damping design essentially transforms the theoretical optimization parameters of the second optimized structural parameter set into an engineering-featured physical structure. The magnetic circuit segmentation design ensures low loss and low noise from two dimensions: magnetic energy loss control and vibration source suppression. The coil damping design further enhances noise control from two aspects: vibration transmission attenuation and structural stability. Both work together to achieve the optimization goals of the second optimized structural parameter set, enabling the synergistic optimization of low loss and low noise to be translated from the parameter level to the physical structure level, ensuring that the optimization results are effectively transformed into the actual performance advantages of the product.

[0035] In this embodiment of the invention, while addressing the issues of synergistic optimization of mechanical strength, compactness, low loss, and low noise in dry-type transformers, stable insulation performance under harsh environments is also a core prerequisite for ensuring long-term reliable operation of the equipment. A targeted approach employing optimized insulation material processes and a synergistic design of a heat dissipation and temperature control system can solidify the foundation of insulation protection from the source and dynamically mitigate the failure risks caused by environmental factors. Specifically, the insulation material selected is H-grade high-temperature resistant epoxy resin. After modification with E-51 (bisphenol A type epoxy resin), this material possesses long-term temperature resistance and resistance to thermal shock, enabling it to adapt to harsh operating conditions of high temperatures and sudden temperature changes, fundamentally improving the environmental tolerance of the insulation layer. The accompanying vacuum impregnation process, through a continuous process of vacuum pre-extraction, pressurized resin injection, and high-temperature curing, can fully remove air and moisture from the windings, core, and other components, ensuring an insulation layer fill rate of no less than 98%, completely eliminating the risk of air bubbles, forming a dense and uniform insulating protective layer, and preventing insulation breakdown caused by the expansion and rupture of air bubbles at high temperatures. The heat dissipation and temperature control system constructs an active protection barrier. Spiral aluminum fins between the core and the outer shell, combined with an axial flow fan, form a highly efficient heat dissipation channel. By increasing the heat dissipation area and using forced ventilation, heat exchange efficiency is improved, keeping the core temperature rise below 40K even in high-temperature environments. The intelligent temperature control system consists of multi-point temperature sensors, a PLC controller, and a variable frequency fan. By collecting real-time temperature data from the windings, core, and outer shell, it achieves graded heat dissipation control. For example, at 70℃, the fan runs at full speed to enhance heat dissipation; at 80℃, a backup heat dissipation branch is activated to ensure cooling, ultimately ensuring the equipment can withstand extreme conditions of 90℃ high temperature and 95% high humidity. Continuous operation without insulation failure in harsh environments: Through the synergy of high-performance insulation materials and refined impregnation processes, the inherent reliability of the insulation layer is improved from both material selection and structural forming aspects, eliminating inherent defects such as bubbles, and significantly enhancing the insulation layer's high-temperature resistance and thermal shock resistance, laying a solid foundation for insulation protection in harsh environments. At the same time, the combined design of efficient heat dissipation channels and intelligent temperature control system enables precise temperature monitoring and dynamic adjustment, effectively controlling the temperature rise of the equipment during operation and avoiding high temperature-induced insulation aging. Furthermore, stable temperature control avoids condensation in high-humidity environments, blocking insulation failure paths from an environmental adaptation perspective.

[0036] The above describes the method for optimizing the structural parameters of a dry-type transformer in the embodiments of the present invention. The following describes the device for optimizing the structural parameters of a dry-type transformer in the embodiments of the present invention. Please refer to [link / reference]. Figure 2 One embodiment of the dry-type transformer structural parameter optimization device of the present invention includes:

[0037] Simulation module 201: used to obtain a first initial structural parameter set, and perform simulation operations based on the first initial structural parameter set to obtain a first set of parameters to be optimized;

[0038] First optimization module 202: used to perform parameter optimization processing on the first set of parameters to be optimized to obtain a first optimized structural parameter set;

[0039] The second optimization module 203 is used to obtain a second set of parameters to be optimized and a preset optimization objective, and to use a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain a second set of optimized structural parameters.

[0040] Parameter integration module 204: used to integrate the first optimized structure parameter set and the second optimized structure parameter set to obtain the target optimized structure parameter set.

[0041] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0042] above Figure 2 The dry-type transformer structural parameter optimization device in this embodiment of the invention is described in detail from the perspective of modular functional entities. The dry-type transformer structural parameter optimization device in this embodiment of the invention is described in detail below from the perspective of hardware processing.

[0043] Figure 3 This is a schematic diagram of a dry-type transformer structural parameter optimization device 300 provided in an embodiment of the present invention. The dry-type transformer structural parameter optimization device 300 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) storing application programs 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the dry-type transformer structural parameter optimization device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute the series of instruction operations in the storage media 330 on the dry-type transformer structural parameter optimization device 300 to implement the steps of the dry-type transformer structural parameter optimization method provided in the above-described method embodiments.

[0044] The dry-type transformer structural parameter optimization device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated dry-type transformer structural parameter optimization device structure does not constitute a limitation on the dry-type transformer structural parameter optimization device, and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0045] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the dry-type transformer structural parameter optimization method.

[0046] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0047] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0048] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the structural parameters of a dry-type transformer, characterized in that, include: An initial three-dimensional simulation model is constructed using simulation tools to obtain a first initial structural parameter set. Simulation operations are performed based on the first initial structural parameter set and the initial three-dimensional simulation model to obtain a first set of parameters to be optimized. The first initial structural parameter set includes winding copper flat wire cross-sectional parameters, core lamination thickness parameters, and winding preload parameters. The first set of parameters to be optimized includes initial electromagnetic force distribution parameters, initial winding deformation parameters, and initial core displacement parameters. The structural weak points of the initial three-dimensional simulation model are determined based on the initial electromagnetic force distribution parameters. Based on the initial winding deformation parameters and the initial core displacement parameters, determine the short-circuit withstand strength non-compliance items of the initial three-dimensional simulation model; Based on the structural weak points and the short-circuit withstand failure, the initial three-dimensional simulation model is optimized to obtain an optimized three-dimensional simulation model. The winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters are input into the optimized three-dimensional simulation model for simulation calculations to obtain optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters. The optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters are then integrated to obtain a first optimized structural parameter set. A second set of parameters to be optimized and a preset optimization objective are obtained. A multi-objective genetic algorithm is used to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain a second set of optimized structural parameters. By integrating the first optimized structural parameter set and the second optimized structural parameter set, the target optimized structural parameter set is obtained.

2. The method for optimizing the structural parameters of a dry-type transformer according to claim 1, characterized in that, The step of performing simulation operations based on the first initial structural parameter set and the initial three-dimensional simulation model to obtain the first parameter set to be optimized includes: The cross-sectional parameters of the winding copper flat wire and the thickness parameters of the iron core lamination are input into the initial three-dimensional simulation model for simulation calculation to obtain the initial electromagnetic force distribution parameters. The initial electromagnetic force distribution parameters and the winding preload parameters are input into the initial three-dimensional simulation model for simulation calculation to obtain the initial winding deformation parameters and the initial core displacement parameters. By integrating the initial electromagnetic force distribution parameters, the initial winding deformation parameters, and the initial core displacement parameters, the first set of parameters to be optimized is obtained.

3. The method for optimizing the structural parameters of a dry-type transformer according to claim 1, characterized in that, The first optimized structural parameter set is obtained by integrating the optimized electromagnetic force distribution parameters, the optimized winding deformation parameters, and the optimized core displacement parameters, including: Obtain the preset winding deformation threshold and the preset volume weight threshold; The optimized winding deformation parameters are compared with the winding deformation threshold. Based on the optimized 3D simulation model, the volume weight parameters are obtained, and the volume weight parameters are compared with the volume weight threshold. When the optimized winding deformation parameter is less than or equal to the winding deformation threshold, and the volumetric weight parameter is less than or equal to the volumetric weight threshold, the optimized electromagnetic force distribution parameter, the optimized winding deformation parameter, and the optimized core displacement parameter are integrated to obtain the first optimized structural parameter set.

4. The method for optimizing the structural parameters of a dry-type transformer according to claim 1, characterized in that, The method employs a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective, thereby obtaining a second optimized structural parameter set, including: The multi-objective genetic algorithm is used to randomly generate multiple initial candidate parameter sets based on the second set of parameters to be optimized; The pre-trained coupled model is invoked to evaluate the performance of multiple sets of initial candidate parameters, thereby obtaining a performance evaluation parameter set corresponding to each set of initial candidate parameters; Based on the optimization objective and multiple sets of performance evaluation parameters, multiple sets of initial candidate parameter sets are screened to obtain multiple sets of high-quality candidate parameter sets; The multi-objective genetic algorithm is used to perform multiple rounds of selection, crossover and mutation processing on multiple sets of high-quality candidate parameters to obtain a set of high-quality candidate parameters that satisfies the optimization objective, and the set of high-quality candidate parameters is used as the second optimization structure parameter set.

5. A device for optimizing the structural parameters of a dry-type transformer, characterized in that, include: Simulation module: Used to construct an initial three-dimensional simulation model using simulation tools, obtain a first initial structural parameter set, and perform simulation operations based on the first initial structural parameter set and the initial three-dimensional simulation model to obtain a first set of parameters to be optimized. The first initial structural parameter set includes winding copper flat wire cross-sectional parameters, core lamination thickness parameters, and winding preload parameters. The first set of parameters to be optimized includes initial electromagnetic force distribution parameters, initial winding deformation parameters, and initial core displacement parameters. First optimization module: used to determine the structural weak points of the initial three-dimensional simulation model based on the initial electromagnetic force distribution parameters; Based on the initial winding deformation parameters and the initial core displacement parameters, determine the short-circuit withstand strength non-compliance items of the initial three-dimensional simulation model; Based on the structural weak points and the short-circuit withstand failure, the initial three-dimensional simulation model is optimized to obtain an optimized three-dimensional simulation model. The winding copper flat wire cross-section parameters, core lamination thickness parameters, and winding preload parameters are input into the optimized three-dimensional simulation model for simulation calculations to obtain optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters. The optimized electromagnetic force distribution parameters, optimized winding deformation parameters, and optimized core displacement parameters are then integrated to obtain a first optimized structural parameter set. The second optimization module is used to obtain a second set of parameters to be optimized and a preset optimization objective, and to use a multi-objective genetic algorithm to perform multiple rounds of iterative optimization on the second set of parameters to be optimized based on the optimization objective to obtain a second optimized structural parameter set. Parameter integration module: used to integrate the first optimized structure parameter set and the second optimized structure parameter set to obtain the target optimized structure parameter set.

6. A device for optimizing the structural parameters of a dry-type transformer, characterized in that, The dry transformer structural parameter optimization device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the dry-type transformer structural parameter optimization device to perform the steps of the dry-type transformer structural parameter optimization method as described in any one of claims 1-4.

7. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the various steps of the dry-type transformer structural parameter optimization method as described in any one of claims 1-4.

Images