Transformer anti-short-circuit capability dynamic checking method and system based on transient-time variation

Abstract

The invention discloses a transient-time varying-based transformer anti-short circuit capability dynamic checking method and system. According to the method, a high-fidelity digital twinborn model is established by constructing a four-layer database comprising a foundation, a material, a structure and operation parameters. And performing transient electromagnetic-structure coupling simulation by using the reconstructed real short-circuit current waveform driving model to obtain transient electromagnetic force, dynamic deformation and stress distribution prediction results under short-circuit impact. Meanwhile, a multi-axis degradation model covering mechanical, electrical and thermal dimensions is constructed, long-term evolution of equipment is quantified, conventional working conditions are converted into equivalent short-circuit impact times, and a comprehensive capability index is calculated. Finally, the transient impact risk and the long-term degradation degree are coupled, a four-quadrant risk assessment matrix is constructed, a comprehensive risk quantitative index is calculated, and dynamic grading and accurate positioning of the overall risk state of the transformer are achieved. According to the method, the problems of isolated island and single risk dimension of traditional static check information are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of transformer safety evaluation, in particular, to a transformer short-circuit resistance dynamic checking method and system based on transient-time-varying. BACKGROUND

[0002] The power transformer is the core hub equipment of the power grid, and its safe and stable operation is directly related to the power supply reliability. The short-circuit current impact on the transformer in operation is one of the main reasons for causing the winding deformation, insulation damage and even sudden damage, which poses a major safety threat to the power grid. However, the traditional static checking method based on factory parameters and design standards cannot accurately reflect the actual state and real-time risk of the equipment after long-term operation, material aging and multiple impacts, and has the disadvantages of "over-conservatism" or "insufficient evaluation".

[0003] The existing transformer short-circuit resistance evaluation technology mainly has the following three significant shortcomings: first, the evaluation is static and isolated. The existing method mainly depends on design drawings and factory test data, and does not consider the dynamic change factors such as actual operation history, material performance degradation, insulation aging and structure loosening after the equipment is put into operation, the evaluation model is disconnected with the actual equipment state, and an "information island" is formed. Second, the risk dimension is single. The traditional method usually only focuses on whether the peak current or electric power under single short-circuit impact exceeds the design threshold, lacks simulation of key mechanical details such as transient electromagnetic force, dynamic deformation process and stress concentration, and does not consider the cumulative mechanical fatigue in long-term operation and the chronic degradation factors such as insulation thermal aging, and the evaluation perspective is one-sided. Third, there is a lack of comprehensive risk quantification and decision support. The existing technology often gives a binary conclusion of "pass" or "not pass", cannot quantify the current comprehensive risk level of the equipment, and cannot locate the multi-dimensional risk combining the transient impact severity and long-term degradation degree, which is difficult to guide the operation and maintenance personnel to develop differentiated preventive measures, and the decision support capability is weak. SUMMARY

[0004] The main purpose of the present application is to provide a transformer short-circuit resistance dynamic checking method and system based on transient-time-varying, to at least solve the problems of information islandization and single risk dimension in the prior art. The essential safety level and asset management lean level of the power grid are improved.

[0005] In order to achieve the above purpose, a transformer short-circuit resistance dynamic checking method and system based on transient-time-varying are provided.

[0006] In a first aspect, the present application provides a transformer short-circuit resistance dynamic checking method based on transient-time-varying, the method comprising:

[0007] A four-layer standardized parameter database of the transformer is constructed, and the four-layer standardized parameter database is processed to construct a digital twin model of the transformer.

[0008] A real short-circuit current waveform of the transformer is collected, and short-circuit current characteristics are extracted from the real short-circuit current waveform. The short-circuit current excitation is input into the digital twin model to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain a transient prediction result. A corresponding transient prediction report is generated according to the prediction result.

[0009] A multi-axis degradation model is constructed to quantify the long-term evolution process of the transformer. The conventional working condition of the transformer is quantified as an equivalent short-circuit impact number. The comprehensive capability index of the transformer is calculated according to the equivalent short-circuit impact number. A long-term degradation prediction model is constructed. The long-term evolution process is evaluated by using the long-term degradation prediction model to obtain a long-term degradation prediction result. A long-term degradation prediction report is generated.

[0010] The transient prediction result and the long-term degradation prediction result are coupled. A four-quadrant risk assessment matrix is constructed, and a comprehensive risk quantification index is calculated. The overall risk state of the transformer is classified according to the comprehensive risk quantification index.

[0011] Specifically, a four-layer standardized parameter database of the transformer is constructed, and the four-layer standardized parameter database is processed to construct a digital twin model of the transformer, including:

[0012] The basic parameters, material parameters, structure parameters and operation parameters of the transformer are collected. The four-layer standardized parameter database of the transformer is constructed by using the basic parameters, material parameters, structure parameters and operation parameters.

[0013] The time-space consistency and physical rule of all data in the four-layer standardized parameter database are verified to obtain a verified four-layer standardized parameter database. The same parameter in the verified four-layer standardized parameter database is fused by using a data fusion algorithm to construct a digital twin model of the transformer.

[0014] Specifically, a real short-circuit current waveform of the transformer is collected, and short-circuit current characteristics are extracted from the real short-circuit current waveform, including:

[0015] The real short-circuit current waveform of the relay protection fault recorder in the transformer substation is extracted. A cutoff frequency is preset, and the real short-circuit current waveform is filtered by using the cutoff frequency. The saturation starting point of the real short-circuit current waveform is identified according to the distortion of the real short-circuit current waveform. The real short-circuit current waveform during the saturation period is reconstructed by using a fitting extrapolation method.

[0016] Extract the short-circuit current characteristics of the reconstructed real short-circuit current waveform, the short-circuit current characteristics including the first large half-wave peak current, the asymmetry degree, the effective action time and the main frequency component.

[0017] Specifically, input the short-circuit current excitation into the digital twin model to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain the transient prediction result, including:

[0018] Extract the three-phase short-circuit current time curve of the relay protection fault recorder as the short-circuit current excitation, and input the short-circuit current excitation into the digital twin model;

[0019] Adopt the sequential coupling solution method to calculate the transient electromagnetic force of the winding in the transformer under the short-circuit current excitation, and calculate the dynamic deformation and stress distribution of the winding by using the transient electromagnetic force to complete the simulation;

[0020] Extract the key mechanical indicators of the winding based on all the data in the simulation process, compare the key mechanical indicators with the yield criterion and the instability criterion to evaluate whether the winding has deformation risk or instability risk, and calculate the short-time risk index under single short-circuit impact, the key mechanical indicators including the maximum equivalent stress, the maximum radial and axial deformation and the stress concentration coefficient;

[0021] Summarize the transient electromagnetic force, the dynamic deformation and stress distribution key mechanical indicators and the short-time risk index to obtain the transient prediction result.

[0022] Specifically, construct a multi-axis degradation model to quantify the long-term evolution process of the transformer, including:

[0023] Align and fuse multi-source time series data by using the operation parameters in the four-layer standardized parameter database, establish a device operation condition time series database, and extract key operation characteristic parameters from the device operation condition time series database, the multi-source time series data including the load rate, the oil temperature, the vibration and the oil chromatogram;

[0024] Construct a multi-axis degradation model according to the key operation characteristic parameters, the multi-axis degradation model including a winding compression force attenuation model based on the pad creep equation and temperature effect in the mechanical axis, an insulation paper polymerization degree attenuation model and an insulation strength correlation model based on the Arrhenius equation in the electrical axis, and a conductor material softening and fatigue characteristic correction model based on the cumulative thermal history in the thermal axis, and quantify the long-term evolution process of the transformer by the multi-axis degradation model.

[0025] Specifically, quantify the conventional operating condition of the transformer as the equivalent short-circuit impact number, and calculate the comprehensive capability index of the transformer according to the equivalent short-circuit impact number, including:

[0026] The types and magnitudes of mechanical stresses generated on windings under various conventional operating conditions are analyzed, and a conversion algorithm from conventional operating conditions to short-circuit impacts is established. Based on the conversion algorithm, various conventional operating conditions are quantified into equivalent short-circuit impact times.

[0027] The total cumulative damage is calculated based on the actual number of short circuits and the equivalent number of short circuit impacts using Miner's linear cumulative damage theory. The key state parameters of the winding are dynamically updated based on the total cumulative damage. The key state parameters include the remaining fatigue life, the current yield strength, and the current clamping force.

[0028] Using the updated key state parameters, the transformer's short-circuit withstand capability is decomposed into three dimensions: resistance to plastic deformation, resistance to instability, and resistance to fatigue fracture. A weighted comprehensive evaluation method is used to calculate a comprehensive capability index to quantify the current remaining short-circuit withstand capability.

[0029] Specifically, a long-term degradation prediction model is constructed, and the long-term evolution process is evaluated using this model to obtain long-term degradation prediction results. A long-term degradation prediction report is then generated, including:

[0030] A long-term degradation prediction model was constructed by combining time series analysis and physical model extrapolation. The long-term degradation prediction model was used to make dynamic predictions of clamping force attenuation, insulation aging and cumulative damage growth under different future operating scenarios to obtain long-term degradation prediction results.

[0031] A long-term degradation forecast report is generated by combining the comprehensive capability index and the long-term degradation forecast results.

[0032] Specifically, transient prediction results are coupled with long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. Based on this comprehensive risk quantification index, the overall risk status of the transformer is classified, including:

[0033] A two-dimensional risk assessment dataset is constructed based on transient and long-term degradation prediction results. A four-quadrant risk assessment matrix is ​​constructed with transient prediction results as the horizontal axis and long-term degradation prediction results as the vertical axis. The matrix is ​​divided into four decision regions: immediate intervention, preventive hardening, enhanced monitoring, and routine operation and maintenance, so as to achieve intuitive positioning of risk status.

[0034] The transient prediction results and the long-term degradation prediction results are coupled through a weighted fusion algorithm to calculate a comprehensive risk quantification index that reflects the overall risk status of the transformer. The comprehensive risk quantification index is then classified into corresponding risk levels according to a preset risk threshold.

[0035] Specifically, the transient prediction results are coupled with the long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. Based on this comprehensive risk quantification index, the overall risk status of the transformer is classified. This also includes:

[0036] The comprehensive risk quantification index of the transformer is mapped to the four-quadrant risk assessment matrix and its corresponding risk area is determined. Based on the comprehensive risk quantification index, risk level and risk area, a comprehensive risk assessment result containing clear risk positioning and classification is generated.

[0037] Secondly, the present invention provides a dynamic verification system for the short-circuit withstand capability of transformers based on transient-time-varying conditions. The system applies the method described in the first part and includes:

[0038] The digital twin model building unit is used to build a four-layer standardized parameter database for transformers and to process the four-layer standardized parameter database to build a digital twin model of the transformer.

[0039] The transient short-circuit impact simulation and prediction unit is connected to the digital twin model construction unit. The transient short-circuit impact simulation and prediction unit is used to acquire real short-circuit current waveforms and extract short-circuit current characteristics. It inputs the short-circuit current characteristics into the digital twin model as short-circuit current excitation to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain transient prediction results. Based on the prediction results, a corresponding transient prediction report is generated.

[0040] The long-term performance degradation modeling and prediction unit is connected to the digital twin model building unit. The long-term performance degradation modeling and prediction unit is used to build a multi-axis degradation model to quantify the long-term evolution process of the transformer, quantify the normal operating conditions into the equivalent short-circuit impact number, calculate the comprehensive capability index based on the equivalent short-circuit impact number, build a long-term degradation prediction model, use the long-term degradation prediction model to evaluate the long-term evolution process to obtain the long-term degradation prediction results, and generate a long-term degradation prediction report.

[0041] The comprehensive risk assessment unit is connected to the long-term performance degradation modeling and prediction unit and the transient short-circuit impact simulation and prediction unit. The comprehensive risk assessment unit is used to couple the transient prediction results with the long-term degradation prediction results, construct a four-quadrant risk assessment matrix and calculate the comprehensive risk quantification index, and classify the overall risk status of the transformer according to the comprehensive risk quantification index.

[0042] This application provides a method and system for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions. This method creates a high-fidelity digital twin model by constructing a four-layer standardized database encompassing foundation, materials, structure, and operating parameters. It utilizes filtered and reconstructed real short-circuit current waveforms as excitation to perform refined simulations of transient electromagnetic forces, dynamic deformation, and stress distribution to assess the risk of a single impact. Simultaneously, the method establishes a long-term multi-axis degradation model covering mechanical, electrical, and thermal aspects, quantifying conventional operating conditions into equivalent short-circuit impact counts. It dynamically calculates the comprehensive capability index using cumulative damage theory and predicts future degradation trends. Finally, by coupling the severity of transient impacts with the degree of long-term degradation, a four-quadrant risk assessment matrix is ​​constructed, and a comprehensive risk quantification index is calculated. This enables dynamic and graded assessment of the overall short-circuit withstand capability of the transformer, providing a complete technical system for shifting from "static verification" to "dynamic risk warning" and precise operation and maintenance decision-making. Attached Figure Description

[0043] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0044] Figure 1 A flowchart illustrating a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions provided in this application;

[0045] Figure 2 A connection diagram of a dynamic verification system for transformer short-circuit withstand capability based on transient-time-varying conditions provided in this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0047] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention 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 embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0048] In this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0049] This application provides a method and system for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions. First, a four-layer standardized database of the transformer's foundation, materials, structure, and operating parameters is established to construct a high-fidelity digital twin model. The method collects and reconstructs short-circuit current waveforms recorded by field relay protection fault recorders, using these waveforms as excitations applied to the digital model. This allows for detailed simulation of the transformer's transient electromagnetic force, dynamic deformation, and stress distribution under impact, assessing the short-term risk under a single impact. Simultaneously, multi-source time-series parameters such as load rate and oil temperature are extracted from operating data to construct a multi-axis degradation model encompassing mechanical axis (compression force attenuation), electrical axis (insulation aging), and thermal axis (material fatigue). This model converts daily operating conditions into equivalent short-circuit impacts and, combined with cumulative damage theory, dynamically calculates a comprehensive capability index characterizing the current remaining short-circuit withstand capability, thereby predicting long-term degradation trends. Finally, by coupling transient impact risk and long-term degradation degree, a four-quadrant risk assessment matrix with the two as the axis is constructed, and a comprehensive risk quantification index is calculated, thereby realizing dynamic classification, intuitive positioning and differentiated operation and maintenance decision support for the overall risk status of transformers.

[0050] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0051] Figure 1 A flowchart illustrating a multimodal and digital twin-based adaptive laser fat decomposition method provided in this application is shown below. Figure 1 As shown in this embodiment, an adaptive laser fat decomposition method based on multimodal and digital twin technologies is provided. The method includes:

[0052] A four-layer standardized parameter database for transformers is constructed, and the four-layer standardized parameter database is processed to construct a digital twin model of the transformer.

[0053] The actual short-circuit current waveform of the transformer is collected and the short-circuit current characteristics are extracted from the actual short-circuit current waveform. The short-circuit current excitation is input into the digital twin model to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain transient prediction results. Based on the prediction results, the corresponding transient prediction report is generated.

[0054] A multi-axis degradation model is constructed to quantify the long-term evolution process of the transformer. The normal operating conditions of the transformer are quantified into the equivalent number of short-circuit impacts. The comprehensive capability index of the transformer is calculated based on the equivalent number of short-circuit impacts. A long-term degradation prediction model is constructed. The long-term degradation prediction model is used to evaluate the long-term evolution process to obtain the long-term degradation prediction results and generate a long-term degradation prediction report.

[0055] The transient prediction results are coupled with the long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. The overall risk status of the transformer is then classified according to the comprehensive risk quantification index.

[0056] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions. The method first constructs a four-layer standardized database integrating transformer foundation, materials, structure, and operating parameters, and then establishes a high-fidelity digital twin model based on this database. Subsequently, the method uses processed real short-circuit current waveforms as input to drive the model in transient electromagnetic-structural coupling simulation to accurately assess the electromagnetic force, dynamic deformation, and stress distribution risk under a single short-circuit impact. Simultaneously, the method also establishes a long-term degradation model covering multiple dimensions such as mechanical, electrical, and thermal aspects, quantifying the transformer's normal operating conditions into equivalent short-circuit impact counts, and calculating a comprehensive capability index reflecting the current health status using cumulative damage theory. Finally, by coupling the severity of transient impacts with the cumulative effects of long-term degradation, a four-quadrant risk assessment matrix is ​​constructed to calculate a quantitative comprehensive risk index, achieving the classification and precise positioning of the transformer's risk status.

[0057] This method represents a fundamental shift in transformer short-circuit withstand capability assessment from a static, isolated "threshold verification" to a dynamic, comprehensive "risk warning." First, by establishing a digital twin model based on multi-source data, it upgrades the assessment object from a static drawing to a dynamic mirror image, significantly improving the realism and accuracy of the simulation. Second, this method, for the first time, systematically incorporates multi-axis (mechanical, electrical, and thermal) degradation effects during long-term operation into short-circuit risk considerations, addressing the critical deficiency of traditional methods that neglect the accumulation of chronic equipment damage. Finally, by coupling transient impacts with long-term degradation and employing a four-quadrant matrix for visualized risk localization, it not only achieves a quantitative assessment of the overall equipment risk but, more importantly, provides a clear and hierarchical scientific basis for operation and maintenance decisions (such as immediate intervention and enhanced monitoring), thereby significantly improving the lean management level of power grid assets and the effectiveness of preventative maintenance.

[0058] Specifically, a four-layer standardized parameter database for the transformer is constructed, and the database is processed to build a digital twin model of the transformer, including:

[0059] Collect basic parameters, material parameters, structural parameters, and operating parameters of the transformer, and use these parameters to construct a four-layer standardized parameter database for the transformer.

[0060] A verified four-layer standardized parameter database is obtained by performing spatiotemporal consistency verification and physical rule verification on all data in the four-layer standardized parameter database. A digital twin model of the transformer is then constructed by weighted fusion of multi-source data of the same parameter in the verified four-layer standardized parameter database using a data fusion algorithm.

[0061] This application provides a dynamic verification method for the short-circuit withstand capability of transformers based on transient-time-varying conditions. This method creates a digital twin model of the transformer by constructing a standardized four-layer data architecture. The specific steps are as follows: Systematically collect basic transformer information (such as model number and nameplate data), material properties (such as conductor and insulation material parameters), precise geometric parameters, and historical and real-time operating status data to form a hierarchical standardized database. Before building the model, all data undergoes rigorous quality control, including verifying the logical consistency of the data in time and space, and performing rule-based verification according to physical laws (such as material strength limits and energy conservation) to eliminate abnormal and contradictory data. Finally, a data fusion algorithm is used to weightedly fuse the same type of parameters from different sources (such as design drawings, measured data, and online monitoring) to form a unified, highly reliable data input, thereby driving the construction of a digital twin model that accurately reflects the true physical state and operating history of a specific transformer.

[0062] Basic parameters are collected from the PMS system, equipment ledger, and test reports: model, capacity, voltage ratio, connection group, and date of manufacture. Material parameters are collected from design drawings and bills of materials: winding conductor grade and mechanical properties, insulation paper type and thermal aging parameters, and pad compression characteristic curve. Structural parameters are collected from manufacturing drawings and installation records: winding geometry, support bar and pad distribution, and axial clamping device parameters. Operating parameters are collected from SCADA and online monitoring systems: three-year load rate curves, temperature records, historical short-circuit counts, and peak current.

[0063] The collected data underwent spatiotemporal consistency verification and physical rule verification. Spatiotemporal consistency verification checked for numerical differences in the same parameter across different sources. Physical rule verification validated the matching of short-circuit impedance and winding dimensions, as well as the rationality of temperature rise and load. A confidence-weighted fusion algorithm was used for multi-source data, with weights allocated as follows: measured value 0.4, online monitoring value 0.3, manually recorded value 0.2, and extrapolated value 0.1.

[0064] A three-dimensional parametric geometric model is established based on the fused parameters. Material properties are assigned hierarchically: conductivity and yield strength of the conductor, BH curve of the core, and dielectric constant of the insulation material. Next, a finite element mesh is created: the winding region is refined (element size ≤ 5mm), while other regions are appropriately coarsened. A blockchain-based evidence index is established to record the source and update time of key parameters.

[0065] This method fundamentally solves the pain points of low data quality and "model inaccuracy" in digital twin model construction. Through a standardized four-layer data structure, it ensures the integrity and systematic nature of the information required by the model, overcoming the drawbacks of fragmented and inconsistent data formats in traditional methods. Its introduced spatiotemporal consistency verification and physical rule verification can automatically identify and correct errors, omissions, or contradictions in the original data, significantly improving the reliability and accuracy of the input data and laying a foundation for the credibility of subsequent simulations. Furthermore, the use of data fusion algorithms to weight and process multi-source data intelligently integrates data sources with different levels of precision and reliability, thereby constructing a high-fidelity twin model that more closely reflects the actual individual differences of the equipment, providing a solid data foundation for subsequent accurate simulation and evaluation.

[0066] Specifically, the actual short-circuit current waveform of the transformer is acquired, and short-circuit current characteristics are extracted from the actual short-circuit current waveform, including:

[0067] Extract the actual short-circuit current waveform from the relay protection fault recorder in the substation where the transformer is located, preset the cutoff frequency and use the cutoff frequency to filter the actual short-circuit current waveform, identify the saturation start point of the actual short-circuit current waveform based on the distortion of the actual short-circuit current waveform, and reconstruct the actual short-circuit current waveform during the saturation period using the fitting extrapolation method.

[0068] The short-circuit current characteristics of the reconstructed real short-circuit current waveform are extracted. These characteristics include the peak current of the first half-wave, asymmetry, effective duration, and main frequency components.

[0069] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions. This method extracts the actual short-circuit current waveform from the relay protection fault recorder of the substation where the transformer is located as the raw input for evaluation. To address the potential electromagnetic transformer saturation distortion problem in the raw waveform, a cutoff frequency is first set to filter the waveform to suppress high-frequency interference. Then, the starting point of current distortion caused by transformer saturation is accurately identified, and a fitting extrapolation algorithm (such as mathematical extrapolation based on pre-saturation waveform characteristics) is used to reconstruct the actual saturation current waveform that is obscured by distortion. After obtaining the complete and accurate reconstructed current waveform, this method systematically extracts a series of key electrical characteristic parameters, including the peak current of the first major half-wave, asymmetry (to characterize the DC component), the effective duration of the short-circuit current, and the main frequency components contained in the waveform, providing high-quality input for subsequent accurate electromagnetic force calculations.

[0070] Three-phase current waveforms were extracted from a fault recorder with a sampling frequency ≥10kHz. A Butterworth low-pass digital filter with a cutoff frequency set to 5kHz was used to denoise the waveforms, eliminating high-frequency sampling noise and carrier interference. Subsequently, the saturation start point of the current transformer was identified based on waveform distortion characteristics (such as flat tops and clipped tops), and a fitting extrapolation algorithm based on the principle of flux linkage conservation was applied to reconstruct the true primary current waveform during saturation. Finally, the three-phase currents were time-referenced and phase-corrected to ensure that the waveforms reflected the true initial electrical angle of the fault. After preprocessing, feature parameters were extracted: the peak current of the first large half-wave was accurately calculated as the core input for electromagnetic force calculation; the asymmetry was determined by the ratio of the first peak to the peak value of subsequent symmetrical waves; the effective action time was defined as the period when the current amplitude was continuously higher than 80% of the peak value; and the waveform spectrum was analyzed using Fast Fourier Transform to extract the main harmonic frequency components other than the power frequency. These feature parameters collectively and comprehensively characterize the mechanical excitation characteristics of the short-circuit current, providing accurate load input for subsequent transient simulations.

[0071] This method directly and accurately assesses actual short-circuit events in the field, overcoming the drawbacks of traditional verification methods that rely on conservative estimations using idealized and standardized theoretical short-circuit currents. Its core technological breakthrough lies in effectively solving the long-standing problem of waveform distortion caused by transformer saturation in engineering. Through "identification-reconstruction" technology, it restores the true current during saturation, significantly improving the fidelity of short-circuit current excitation data. Based on the feature parameters extracted from this high-fidelity waveform, it can more realistically reflect the transient severity (such as peak value and asymmetric impact force) and dynamic characteristics (such as frequency components) of short-circuit impacts. This makes the subsequent electromagnetic force simulation and mechanical response analysis results in the digital twin model more accurate and reliable, providing a solid and trustworthy data foundation for assessing the short-circuit withstand capability of transformers.

[0072] Specifically, a short-circuit current excitation is input into the digital twin model to simulate the transient electromagnetic force, dynamic deformation, and stress distribution of the transformer under short-circuit impact, thereby obtaining transient prediction results, including:

[0073] The three-phase short-circuit current time history curve of the relay protection fault recorder is extracted as the short-circuit current excitation, and the short-circuit current excitation is input into the digital twin model.

[0074] The transient electromagnetic force of the winding in a transformer under short-circuit current excitation is calculated using the sequential coupling solution method. The dynamic deformation and stress distribution of the winding are then calculated using the transient electromagnetic force to complete the simulation.

[0075] Based on all data from the simulation process, the key mechanical indicators of the winding are extracted. The winding is evaluated for deformation risk or instability risk by comparing the key mechanical indicators with the yield criterion and the instability criterion. The short-time risk index under a single short-circuit impact is calculated. The key mechanical indicators include the maximum equivalent stress, the maximum radial and axial deformation, and the stress concentration factor.

[0076] Transient prediction results are obtained by summarizing key mechanical indicators of transient electromagnetic force, dynamic deformation and stress distribution, as well as short-term risk index.

[0077] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions. This method utilizes a digital twin model of the transformer, inputting a three-phase short-circuit current time history curve accurately reconstructed from a field fault recorder as dynamic excitation, and performing transient electromagnetic-structural coupling simulation. It employs a sequential coupling solution strategy, first accurately calculating the transient electromagnetic force generated by each turn of the winding under the dynamic current excitation, and then applying this force as a load to the winding structure model to solve for its dynamic deformation and internal stress distribution during the impact process. After simulation, the system automatically extracts key mechanical performance indicators, including maximum equivalent stress, maximum radial and axial deformation, and stress concentration factor. Finally, these indicators are compared with physical criteria such as the material's yield strength and the structural buckling instability threshold to quantitatively assess whether the winding exhibits plastic deformation or instability risk, and to calculate a quantified single-impact short-time risk index. By summarizing all simulation data, mechanical indicators, and risk indices, a transient prediction result report for the specific short-circuit event is generated.

[0078] First, the pre-processed three-phase short-circuit current time history curves are... As an excitation source, the input is fed into a digital twin model of a transformer built on ANSYS Maxwell. The Lorentz force density distribution on the winding is calculated using a transient solver in 50-microsecond steps. The calculation formula is as follows:

[0079]

[0080] in, Represents spatial location ,time The electromagnetic force density vector at that location, Represents spatial location ,time The current density vector at that location, Represents spatial location ,time The magnetic flux density vector at a given location represents the vector cross product operator.

[0081] Subsequently, the electromagnetic force field was mapped as a load onto the structural model of ANSYS Mechanical. Considering the plasticity of the winding material, the nonlinearity of the pad contact, and the boundary conditions of axial preload, an explicit dynamic algorithm was used to solve for the transient displacement of the winding. With stress The response and calculation process are as follows:

[0082] Given initial displacement ,speed For each The resultant force at the nodes is calculated using the following formula:

[0083]

[0084] The formula for calculating acceleration is:

[0085]

[0086] The updated velocity and displacement are calculated using the following formula:

[0087] ;

[0088] ;

[0089] The formula for calculating the strain increment is as follows:

[0090]

[0091] The updated stress is calculated using the following formula:

[0092]

[0093] in, Representative node In time The resultant force vector at that point Representative node In time External load vector at the location, Representative node In time The internal resistance vector at that location, Represents time Cauchy stress tensor at the location. Representative node In time The acceleration vector at that point, Representative node The lumped mass matrix, Representative node In time The velocity vector at that point Representative node In time Displacement vector at that point The strain increment tensor represents the current time step. Represents the strain-displacement transformation matrix. The vector representing the node displacement increment at the current time step. Represents the updated Cauchy stress tensor. This represents the elasticity matrix of the material.

[0094] If the stress exceeds the failure criterion, the marked element fails. Output the transient displacement and stress at the specified time.

[0095] The maximum equivalent stress is extracted from the simulation results, and the calculation formula is as follows:

[0096]

[0097] The formula for calculating the maximum radial deformation is:

[0098]

[0099] The short-term risk index is calculated based on the yield criterion and the instability criterion. The calculation formula is as follows:

[0100]

[0101] in, Represents the maximum equivalent stress. Represents spatial location ,time Equivalent stress at the point, This represents the maximum radial deformation. Represents spatial location ,time The radial displacement vector at that location. Represents a short-term risk index. This represents the current yield strength of the winding conductor material. This represents the maximum radial resultant force calculated by simulation. This represents the critical buckling load of the winding. This represents the maximum permissible radial deformation. This represents the asymmetric shock correction factor.

[0102] Finally, risk classification is completed based on the short-term risk index, and an assessment report is generated. When the winding is within the elastic safety range, it is considered low-risk; when At this time, the winding may enter plastic or near-instability, requiring close monitoring and classified as medium risk; when At this time, the winding has already undergone or is very likely to undergo permanent deformation or instability, requiring immediate intervention, which is a high-risk situation.

[0103] This method enables a refined and dynamic understanding of the complex internal physical processes of transformers under short-circuit impacts. It directly correlates the dynamic time-domain characteristics of the short-circuit current (such as peak value and asymmetry) with the dynamic mechanical response of the winding structure. Through transient coupling simulation, it can not only calculate the peak electromagnetic force but also capture the pulsation and oscillation of the electromagnetic force during the impact, as well as the resulting dynamic deformation accumulation and stress redistribution. This is more accurate and comprehensive than traditional static or quasi-static analysis methods. By introducing physical failure criteria such as yielding and instability and calculating risk indices, it transforms the massive amounts of simulation output data (force, deformation, stress) into intuitive and quantifiable "risk" information, directly serving engineering judgments. This provides unprecedented depth and scientific rigor for assessing the severity of a single short-circuit impact and locating potential weaknesses, representing a significant advancement over traditional peak current-based verification methods.

[0104] Specifically, a multi-axis degradation model is constructed to quantify the long-term evolution of the transformer, including:

[0105] By aligning and fusing operating parameters from a four-layer standardized parameter database with multi-source time-series data, a time-series database of equipment operating conditions is established. Key operating characteristic parameters are extracted from this database. The multi-source time-series data includes load rate, oil temperature, vibration, and oil chromatography.

[0106] A multi-axis degradation model is constructed based on key operating characteristic parameters. The multi-axis degradation model includes a winding clamping force attenuation model based on the pad creep equation and temperature effect for the mechanical axis, an insulation paper polymerization degree attenuation model and insulation strength correlation model based on the Arrhenius equation for the electrical axis, and a conductor material softening and fatigue characteristic correction model based on cumulative thermal history for the thermal axis. The long-term evolution process of the transformer is quantified through the multi-axis degradation model.

[0107] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions. This method first utilizes operational data from a four-layer standardized parameter database to align and fuse multi-source time-series monitoring data such as load rate, oil temperature, vibration, and oil chromatography, forming a unified equipment operating condition time-series database, from which key characteristic parameters are extracted. Based on these characteristic parameters reflecting the actual operating state, this method constructs a "multi-axis degradation model" encompassing mechanical, electrical, and thermal dimensions. In terms of the mechanical axis, the model simulates the attenuation of winding clamping force caused by the creep of the insulating pad under temperature and pressure; in terms of the electrical axis, it simulates the decrease in the degree of polymerization of the insulating paper and its correlation with insulation strength based on the Arrhenius chemical reaction kinetic equation; and in terms of the thermal axis, it considers the softening of conductor material mechanical properties (such as yield strength) and the accumulation of fatigue damage caused by accumulated thermal history. Through this comprehensive model, the complex long-term evolution process of the transformer, consisting of aging, fatigue, and deterioration, can be systematically and quantitatively described.

[0108] First, time-series data on transformer operating conditions, including load rate, top oil temperature, hot spot temperature (calculated value), ambient temperature, winding vibration spectrum, and oil chromatographic gas content, for at least three years are obtained from the SCADA system and online monitoring devices. This data is then precisely aligned and fused using a unified timestamp (e.g., 15-minute intervals) to construct a complete time-series database of equipment operating conditions. Next, feature engineering calculations are performed: based on the historical hot spot temperature, the cumulative thermal aging factor is calculated using the Arrhenius relative aging rate formula, expressed as the number of hours of operation equivalent to a reference temperature (e.g., 110℃). From the vibration monitoring data, the amplitude decay trend of characteristic frequencies (e.g., the first-order axial natural frequency) closely related to the winding and clamping structure status is extracted, and the mechanical stress factor is calculated, typically defined as the ratio of the current amplitude to the initial amplitude during operation. From the load curve, the duration exceeding 80% of the rated load, load fluctuation rate, and number of start-stop cycles are statistically analyzed to comprehensively calculate the electrical stress factor, reflecting the severity of the electrical load. All extracted feature parameters are normalized and stored in a feature library, providing standardized input for subsequent degradation modeling.

[0109] A multi-axis degradation model is constructed, consisting of three interrelated axis systems: a mechanical axis focusing winding clamping system, and a nonlinear creep equation for the compressive strain of the pad material as a function of time and temperature, based on creep test data of the pad material. The equation is as follows:

[0110]

[0111] in, Represents the pad in time ,temperature Total creep strain under the following conditions This represents the constant compressive stress acting on the pad. This represents the initial elastic modulus of the pad material. Represents the creep coefficient. Represents the stress index, Represents loading time. Represents a time index. Represents the creep activation energy. Represents the universal gas constant. This represents the hot spot temperature during transformer operation.

[0112] Furthermore, a model was derived showing that the overall axial clamping force of the winding decreases exponentially with operating time, as shown in the formula:

[0113]

[0114] in, Represents running time The current axial clamping force after that, Represents the initial clamping force. This represents the coefficient of clamping force attenuation. This represents the cumulative running time.

[0115] The electrical axis focuses on solid insulation, using the classic Arrhenius thermal aging model to describe the decay of the degree of polymerization (DP value) of the insulating paper. The formula is:

[0116]

[0117] in, Represents running time The current degree of polymerization of the insulating paper after the insulation paper Represents the initial degree of polymerization of the insulating paper. The reaction rate constant representing aging, , Represents the pre-exponential factor. Represents the activation energy of the reaction. Represents the universal gas constant. The absolute hot spot temperature of the insulating paper is used to establish an empirical correlation model between the DP value and the mechanical strength and breakdown voltage of the insulating paper.

[0118] Focusing on thermal axis conductor materials, based on cumulative thermal history (using a modified Manson-Haferd model) and fatigue damage theory, the yield strength and SN fatigue curve of copper conductors are modified. The formula for the modified yield strength of copper conductors is as follows:

[0119]

[0120] in, Represents running time The current yield strength of the conductor material after that, Represents the initial yield strength of the conductor material. Represents the strength softening coefficient. Represents the cumulative thermal damage function. This represents the temperature-dependent softening function.

[0121] The mechanical axis model quantifies the exponential decay of the winding axial clamping force over operating time using the pad creep equation and temperature effect function. The electrical axis model, based on the Arrhenius equation, quantifies the decay trajectory of the insulation paper's degree of polymerization with hot spot temperature and time, and establishes a quantitative empirical correlation between the degree of polymerization and the mechanical and electrical strength of the insulation, thus transforming chemical aging into calculable performance indicators. The thermal axis model quantifies the softening process of the conductor material's yield strength and fatigue characteristics through a cumulative thermal damage function. These three axis models are coupled through shared time variables and temperature fields, enabling comprehensive calculation of the continuous decay curves of core indicators of the transformer winding's short-circuit withstand capability (such as critical instability load and allowable material stress) under the combined effects of long-term operation and thermal cycling. This achieves a shift from qualitative aging description to quantitative performance prediction, providing a dynamic evolutionary mathematical basis for accurately assessing its remaining short-circuit withstand capability.

[0122] This method breaks through the limitations of traditional approaches that typically focus only on electrical performance aging or single-dimensional degradation. For the first time, it systematically models the long-term degradation of transformers from three interconnected and mutually influential physical dimensions (axis): mechanical, electrical, and thermal. It no longer treats insulation aging, mechanical loosening, and material fatigue as independent events, but rather dynamically correlates them with observable operating parameters (such as temperature and load) through physicochemical models (e.g., the Arrhenius equation and creep equation), achieving a mechanism-driven quantitative assessment of transformer health. This multi-axis integrated model more realistically reflects the actual degradation process of equipment under complex operating conditions, laying a solid theoretical foundation for accurately predicting remaining life and scientifically assessing the actual short-circuit withstand capability in the current state (rather than the factory design capability), thus shifting the assessment of long-term transformer risk from empirical speculation to scientific calculation.

[0123] Specifically, the normal operating conditions of a transformer are quantified into the equivalent number of short-circuit impacts. The comprehensive capability index of the transformer is calculated based on the equivalent number of short-circuit impacts, including:

[0124] The types and magnitudes of mechanical stresses generated on windings under various conventional operating conditions are analyzed, and a conversion algorithm from conventional operating conditions to short-circuit impacts is established. Based on the conversion algorithm, various conventional operating conditions are quantified into equivalent short-circuit impact times.

[0125] The total cumulative damage is calculated based on the actual number of short circuits and the equivalent number of short circuit impacts using Miner's linear cumulative damage theory. The key state parameters of the winding are dynamically updated based on the total cumulative damage. The key state parameters include the remaining fatigue life, the current yield strength, and the current clamping force.

[0126] Using the updated key state parameters, the transformer's short-circuit withstand capability is decomposed into three dimensions: resistance to plastic deformation, resistance to instability, and resistance to fatigue fracture. A weighted comprehensive evaluation method is used to calculate a comprehensive capability index to quantify the current remaining short-circuit withstand capability.

[0127] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions. This method aims to dynamically assess the current true short-circuit withstand capability of a transformer. Its core is the establishment of a "damage equivalence" mechanism. This involves analyzing the periodic mechanical stress induced in the windings by factors such as load changes and temperature cycling during daily operation, and comparing the fatigue damage to the material caused by this stress with the damage caused by a single standard short-circuit impact. This allows for the construction of an algorithm that converts long-term routine operating conditions into "equivalent short-circuit impact counts." Based on this, the method incorporates both the historical actual short-circuit counts and the calculated equivalent impact counts into a framework based on Miner's linear cumulative damage theory to calculate the total cumulative damage degree borne by the winding structure. This damage degree is used to dynamically and in real-time update key parameters characterizing the winding health status, such as remaining fatigue life, current yield strength of the material, and current actual winding clamping force. Finally, using these updated key parameters that reflect the true aging state of the equipment, the complex attribute of transformer short-circuit withstand capability is decomposed into three specific dimensions: "resistance to plastic deformation", "resistance to instability" and "resistance to fatigue fracture". Through a weighted comprehensive evaluation method, a quantitative "comprehensive capability index" is calculated to intuitively characterize the current remaining capability of the transformer after long-term service and multiple impacts.

[0128] First, based on finite element analysis and materials mechanics, a stress mapping relationship between different operating conditions and short-circuit conditions is established: the thermal expansion and contraction of the winding caused by daily / seasonal oil temperature changes is converted into an equivalent axial alternating stress amplitude using the material's thermal expansion coefficient; the electromagnetic force changes caused by load variations are converted into an equivalent radial alternating stress amplitude; and the long-term background vibration is converted into an equivalent vibration stress amplitude through spectral analysis. Then, based on Miner's linear cumulative damage rule, the stress amplitude of each operating condition is compared with the stress amplitude of a reference short-circuit impact using the material's SN curve (stress amplitude-life curve). The damage ratio of a single event is calculated, and the results are accumulated according to the number of occurrences to obtain the total equivalent number of short-circuit impacts. The calculation formula is as follows:

[0129]

[0130] in, Represents the total number of equivalent short-circuit impacts. Represents the working condition type index. Representing the The amplitude of alternating stress caused by this type of operating condition at critical locations in the winding (such as the outermost layer of the conductor). Represents the reference stress amplitude. The inverse slope of the SN (stress-life) curve representing the material. This represents the number of times the first type of operating condition actually occurs within the cycle.

[0131] Finally, the total cumulative damage is calculated by combining the actual number of short circuits and the total number of equivalent short circuit impacts, using the following formula:

[0132]

[0133] in, Represents the total cumulative damage. This represents the total number of short-circuit impacts required to cause fatigue failure of the winding (such as wire breakage). This represents the actual number of short-circuit faults that have occurred in history.

[0134] The critical capability parameters are reduced using the cumulative damage level, calculated using the following formula:

[0135] ;

[0136] ;

[0137] in, This represents the current yield strength of the conductor after considering cumulative damage. Represents the initial yield strength of the conductor. Represents the strength reduction factor. This represents the current critical buckling load of the winding after considering the decay of the clamping force. This represents the initial critical buckling load of the winding under the initial clamping force. This represents the initial axial clamping force. This represents the sensitivity index of the instability load to the clamping force.

[0138] Current insulation withstand strength according to Then, a comprehensive capability index is constructed, which is a weighted sum of three sub-capability indices: the resistance to plastic deformation capability index, the resistance to instability capability index, and the resistance to fatigue fracture capability index. The calculation formula is as follows:

[0139] ;

[0140] ;

[0141] ;

[0142] ;

[0143] in, Represents the index of resistance to plastic deformation. Represents the index of resistance to instability. This represents the fatigue fracture resistance index. , and These represent the weighting coefficients for the resistance to plastic deformation, the resistance to instability, and the resistance to fatigue fracture, respectively.

[0144] The closer the comprehensive capacity index value is to 1, the closer the remaining capacity is to a newly commissioned state.

[0145] This method creatively unifies the transformer's long, continuous routine service history and discrete short-circuit impact history into a quantifiable and cumulative "damage" index, thereby dynamically assessing the equipment's "remaining capacity." It updates the assumption implicit in traditional static verification that "equipment is constant from the factory," to the scientific understanding that "equipment condition dynamically evolves with service history." By introducing equivalent impact conversion and cumulative damage theory, this method places chronic damage from daily operation (such as thermal fatigue) and acute damage caused by sudden short circuits under the same evaluation scale, solving the problem of the separation between the two in previous assessments. The final calculated "comprehensive capacity index" is no longer based on the design value on drawings, but on the "current net value" based on the actual service history of the equipment. This provides a direct, quantitative scientific basis for judging whether an old transformer can still withstand the next short-circuit impact, greatly improving the accuracy of condition assessment and risk warning.

[0146] Specifically, a long-term degradation prediction model is constructed, and the long-term evolution process is evaluated using this model to obtain long-term degradation prediction results. A long-term degradation prediction report is then generated, including:

[0147] A long-term degradation prediction model was constructed by combining time series analysis and physical model extrapolation. The long-term degradation prediction model was used to make dynamic predictions of clamping force attenuation, insulation aging and cumulative damage growth under different future operating scenarios to obtain long-term degradation prediction results.

[0148] A long-term degradation forecast report is generated by combining the comprehensive capability index and the long-term degradation forecast results.

[0149] This application provides a dynamic verification method for transformer short-circuit withstand capability based on transient-time-varying conditions, aiming to make forward-looking predictions of the long-term performance evolution of transformers. Specifically, it constructs a hybrid prediction model that combines time-series analysis based on historical operating data with mechanistic extrapolation methods based on physicochemical laws (such as creep, thermal aging, and fatigue damage). Using this integrated long-term degradation prediction model, it can dynamically simulate and predict the further attenuation trend of winding clamping force, the continuous aging process of insulation materials, and the growth path of cumulative damage for different possible future operating scenarios (e.g., predicted load growth, planned maintenance arrangements, different ambient temperature patterns, etc.), thereby obtaining quantitative, multi-scenario long-term degradation prediction results. Finally, this method integrates a comprehensive capability index reflecting the current state with long-term prediction results revealing future evolution trends to generate a long-term degradation prediction report that includes current status assessment, future risk prediction, and evolution path analysis.

[0150] First, a predictive model is constructed: The degradation model (mechanical, electrical, and thermal) is combined with the equipment's projected future operating load curve (obtainable from power grid planning data or by setting benchmarks, heavy loads, and other typical scenarios) and environmental temperature predictions, serving as input to drive the model for time extrapolation. The prediction process employs a numerical integration method, dynamically calculating the clamping force, cohesion, cumulative damage, and corresponding comprehensive capability index at each future time point in monthly or yearly increments. Next, multi-scenario analysis is conducted, typically including at least a "baseline scenario" (continuing historical average operating conditions), a "heavy load scenario" (load increased by 15-25%), and an "accelerated aging scenario" (considering higher ambient temperatures or more frequent start-stop cycles). Based on the prediction curves, the time points when the comprehensive capability index drops to the preset warning threshold (e.g., 0.7) and critical threshold (e.g., 0.5) are determined. Finally, a long-term risk assessment report is automatically generated. The report is structured and includes: a summary of the current status (current comprehensive capability index and level); historical and future prediction curves for key parameters (tightening force, DP value); graphs showing the remaining capability index changing over time under multiple scenarios, highlighting warning and critical time points; the long-term risk level based on prediction (low, medium, high); and specific, actionable maintenance intervention recommendations and their recommended implementation time windows (e.g., "It is recommended to arrange winding tightening force restoration treatment before xx years ago").

[0151] This method represents a leap from "current status assessment" to "future prediction," providing crucial forward-looking decision-making for asset management and maintenance planning. By integrating data-driven and physical mechanism-based approaches, it leverages the statistical patterns of historical data while respecting the fundamental scientific principles of material aging, thus enhancing the reliability and adaptability of predictions. Capable of dynamic predictions across multiple scenarios, it allows decision-makers to assess the impact of different operating strategies (such as adjusting load and improving cooling) on ​​equipment lifespan and risks, providing a powerful "digital sandbox" tool for optimizing operating methods and developing preventative maintenance plans. The resulting prediction report concretizes abstract "future risks" into understandable quantitative indicators and evolution curves, providing a scientific basis for developing differentiated and cost-effective equipment lifecycle management strategies. It is a core technological support for moving from reactive maintenance to proactive health management.

[0152] Specifically, transient prediction results are coupled with long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. Based on this comprehensive risk quantification index, the overall risk status of the transformer is classified, including:

[0153] A two-dimensional risk assessment dataset is constructed based on transient and long-term degradation prediction results. A four-quadrant risk assessment matrix is ​​constructed with transient prediction results as the horizontal axis and long-term degradation prediction results as the vertical axis. The matrix is ​​divided into four decision regions: immediate intervention, preventive hardening, enhanced monitoring, and routine operation and maintenance, so as to achieve intuitive positioning of risk status.

[0154] The transient prediction results and the long-term degradation prediction results are coupled through a weighted fusion algorithm to calculate a comprehensive risk quantification index that reflects the overall risk status of the transformer. The comprehensive risk quantification index is then classified into corresponding risk levels according to a preset risk threshold.

[0155] This application provides a method and system for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions. This method couples two different time-scale risks faced by the transformer—the severity of transient short-circuit impacts and the cumulative degree of long-term performance degradation—to construct a holistic risk assessment system. Its core is the establishment of a two-dimensional, four-quadrant risk assessment matrix: the transient prediction result (representing the threat level of the current short-circuit impact) is used as the horizontal axis, and the long-term degradation prediction result (representing the degradation level of the equipment's own health) is used as the vertical axis. This matrix clearly divides four decision-making regions, such as the "immediate intervention" region (high impact, high degradation), the "preventive reinforcement" region (low impact, high degradation), the "enhanced monitoring" region (high impact, low degradation), and the "routine operation and maintenance" region (low impact, low degradation), thereby achieving intuitive positioning and classification of risk status. Simultaneously, this method uses a weighted fusion algorithm to calculate a unified, numerical comprehensive risk quantification index from the prediction results of these two dimensions, and classifies this index according to preset thresholds, thereby achieving a comprehensive qualitative and quantitative assessment of the overall risk status of the transformer.

[0156] First, the core outputs are extracted from the transient prediction results, including the short-term risk index, the coordinates of the maximum stress location, and the deformation exceeding the standard indicator. The short-term risk index is then normalized to obtain a standardized short-term risk index with values ​​ranging from [0,1]. Simultaneously, the current remaining short-circuit withstand capability index and predicted remaining life are extracted from the long-term degradation prediction results and converted into a long-term risk index. Finally, these standardized data items, along with the transformer's metadata (such as voltage level, years of operation, and recent maintenance records), are combined to construct a structured JSON-formatted two-dimensional risk assessment dataset, serving as the sole data source for all subsequent analyses and visualizations.

[0157] First, thresholds for high and low risk were set for the standardized short-term and long-term risk indices, respectively. Then, a two-dimensional plane was constructed with the standardized short-term risk index as the horizontal axis and the long-term risk index as the vertical axis. Using the aforementioned thresholds, this plane was divided into four distinct quadrants: the first quadrant was defined as the "immediate intervention zone"; the second quadrant as the "preventive reinforcement zone"; the third quadrant as the "enhanced monitoring zone"; and the fourth quadrant as the "routine operation and maintenance zone." Finally, a mapping algorithm was used to precisely locate the (short-term risk index, long-term risk index) data points of each transformer within this two-dimensional matrix, thereby intuitively determining the current risk zone of the equipment and the corresponding macro-level response strategy.

[0158] Based on the transformer's voltage level (110kV, 220kV, 500kV), a differentiated weighting strategy is adopted to determine the contribution ratio of short-term risk and long-term risk in the comprehensive index. Subsequently, a comprehensive risk quantification index with a value range between [0,1] is calculated using a weighted fusion formula. Finally, based on the three-level risk thresholds set for different voltage levels (for 220kV transformers, the threshold is set when...),... When, it is low risk; when At that time, it was considered a medium risk; when When the risk level is high, the comprehensive risk quantification index of each transformer is divided into specific risk levels (low, medium, high), and the confidence level of the assessment results is calculated, thus completing the transformation from two-dimensional regional positioning to one-dimensional quantitative classification.

[0159] This method fundamentally solves the problem of separating "acute risks" from "chronic risks" in traditional assessments. By using a four-quadrant matrix to visualize and structure complex risk situations, it enables maintenance personnel to clearly understand the nature of the risk (whether it's impact-driven or aging-driven) and quickly pinpoint appropriate management strategies (such as immediate intervention or enhanced monitoring). This coupled assessment avoids misjudgments that can occur with single indicators (e.g., focusing only on a single major impact while ignoring the severe aging of the equipment itself, or vice versa). The weighted fusion to generate a comprehensive risk quantification index provides a unified and operational quantitative basis for comparing risks between different equipment, tracking risk trends, and developing differentiated maintenance resource allocation schemes based on risk levels. This represents a crucial leap in risk assessment from "descriptive analysis" to "guiding decision-making," and is a core tool for improving the lean management of power grid assets.

[0160] Specifically, the transient prediction results are coupled with the long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. Based on this comprehensive risk quantification index, the overall risk status of the transformer is classified. This also includes:

[0161] The comprehensive risk quantification index of the transformer is mapped to the four-quadrant risk assessment matrix and its corresponding risk area is determined. Based on the comprehensive risk quantification index, risk level and risk area, a comprehensive risk assessment result containing clear risk positioning and classification is generated.

[0162] This application provides a method and system for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions. This method calculates the comprehensive risk quantification index of the transformer and maps it to a pre-constructed four-quadrant risk matrix, then further integrates multi-dimensional assessment information to generate a final decision. Specifically, the system first accurately locates the coordinate point (determined by transient and long-term degradation risk values) corresponding to the comprehensive risk quantification index within the four-quadrant matrix, clarifying which specific risk area it belongs to among "immediate intervention," "preventive reinforcement," "enhanced monitoring," or "routine operation and maintenance." Subsequently, it integrates three key pieces of information: the quantified risk index value, the risk level classified according to thresholds (e.g., high risk, medium risk), and the qualitative risk area it occupies, automatically generating a comprehensive risk assessment report containing clear risk location, risk level judgment, and corresponding operation and maintenance strategy recommendations. This report not only provides a risk score but also clearly indicates the source and nature of the risk and the priority actions to be taken.

[0163] The system automatically integrates all intermediate results from the transient and long-term degradation predictions, including basic equipment information, dual-dimensional risk data points, four-quadrant positioning, comprehensive risk indicators and levels, and populates them according to a preset template to generate a complete text report. The report includes a summary of the assessment results, detailed risk analysis, risk positioning visualization charts, and specific improvement recommendations based on the risk level and quadrant. Simultaneously, the system generates structured data in JSON format for inter-system interaction and creates interactive visualization pages containing elements such as four-quadrant scatter plots and risk radar charts to enhance readability and usability. Ultimately, this comprehensive assessment report, containing quantitative indicators, qualitative conclusions, and clear action recommendations, will serve as the core decision-making basis for guiding subsequent transformer operation, maintenance, or technical upgrades.

[0164] This method transforms complex assessment conclusions into clear, direct, and actionable guidelines. By mapping comprehensive indicators back to a four-quadrant matrix and associating them with specific regions, the system not only informs operations and maintenance personnel of the level of risk (quantified level) but also intuitively demonstrates "why it's high" (whether due to recent shocks or long-term aging) and "what should be done" (recommended strategies for the corresponding region). This output method significantly lowers the barrier to understanding and application costs of professional assessment results, making the path from technical analysis to management decision-making extremely clear and efficient. It effectively supports differentiated and precise scheduling of operations and maintenance resources, achieves closed-loop risk management, and improves the intelligence level and response efficiency of power grid asset management.

[0165] Figure 2 A schematic diagram of a dynamic verification system for transformer short-circuit withstand capability based on transient-time-varying conditions provided in this application is shown below. Figure 2As shown, this embodiment provides a dynamic verification system for the short-circuit withstand capability of a transformer based on transient-time-varying conditions. The system includes:

[0166] The digital twin model building unit is used to build a four-layer standardized parameter database for transformers and to process the four-layer standardized parameter database to build a digital twin model of the transformer.

[0167] The transient short-circuit impact simulation and prediction unit is connected to the digital twin model construction unit. The transient short-circuit impact simulation and prediction unit is used to acquire real short-circuit current waveforms and extract short-circuit current characteristics. It inputs the short-circuit current characteristics into the digital twin model as short-circuit current excitation to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain transient prediction results. Based on the prediction results, a corresponding transient prediction report is generated.

[0168] The long-term performance degradation modeling and prediction unit is connected to the digital twin model building unit. The long-term performance degradation modeling and prediction unit is used to build a multi-axis degradation model to quantify the long-term evolution process of the transformer, quantify the normal operating conditions into the equivalent short-circuit impact number, calculate the comprehensive capability index based on the equivalent short-circuit impact number, build a long-term degradation prediction model, use the long-term degradation prediction model to evaluate the long-term evolution process to obtain the long-term degradation prediction results, and generate a long-term degradation prediction report.

[0169] The comprehensive risk assessment unit is connected to the long-term performance degradation modeling and prediction unit and the transient short-circuit impact simulation and prediction unit. The comprehensive risk assessment unit is used to couple the transient prediction results with the long-term degradation prediction results, construct a four-quadrant risk assessment matrix and calculate the comprehensive risk quantification index, and classify the overall risk status of the transformer according to the comprehensive risk quantification index.

[0170] This application provides a method and system for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions. This system is an intelligent software platform integrating four major functions: digital twin construction, transient impact simulation, long-term degradation modeling, and comprehensive risk assessment. Specifically, the system first uses a digital twin model construction unit to collect and process the transformer's foundation, materials, structure, and operating parameters to establish a high-fidelity virtual model. The transient short-circuit impact simulation and prediction unit is responsible for accessing field fault recording data and driving the twin model to perform transient electromagnetic-structural coupling simulation, evaluating the detailed mechanical response and short-term risk under a single short-circuit impact. The long-term performance degradation modeling and prediction unit, based on historical operating data, constructs a multi-axis (mechanical, electrical, and thermal) degradation model, quantifies the health degradation process of the equipment, and predicts its future evolution trend. Finally, the comprehensive risk assessment unit couples the outputs of the first two units to construct a four-quadrant risk assessment matrix, calculates a quantitative comprehensive risk index, and accurately classifies the overall risk status of the transformer, thereby generating a comprehensive risk assessment report and operation and maintenance decision recommendations.

[0171] This system realizes an end-to-end, fully automated closed-loop process for transformer risk assessment, encompassing data acquisition, model building, simulation analysis, and decision support. It encapsulates complex multiphysics simulation, multi-axis degradation mechanism analysis, and big data fusion technology within an integrated software architecture, significantly enhancing the professionalism, efficiency, and repeatability of the assessment. The various functional units work collaboratively, with the digital twin model serving as a unified digital foundation, ensuring consistency and comparability between transient simulation and long-term forecast data. Ultimately, the system not only outputs detailed technical analysis reports but also, through a four-quadrant matrix and comprehensive risk indicators, directly transforms technical conclusions into clear risk levels and operational strategy instructions usable by management. This marks a significant step forward in transformer risk assessment, moving from a discrete, manual operation model reliant on expert experience to a systematic and intelligent decision support system based on models and data, holding immense value for promoting the digital transformation of power asset management.

Claims

1. A dynamic verification method for the short-circuit withstand capability of transformers based on transient-time-varying conditions, characterized in that, The verification method includes: A four-layer standardized parameter database for the transformer is constructed, and the four-layer standardized parameter database is processed to construct a digital twin model of the transformer. The actual short-circuit current waveform of the transformer is collected and the short-circuit current characteristics are extracted from the actual short-circuit current waveform. The short-circuit current excitation is input into the digital twin model to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under short-circuit impact to obtain transient prediction results. A corresponding transient prediction report is generated based on the prediction results. A multi-axis degradation model is constructed to quantify the long-term evolution process of the transformer. The normal operating conditions of the transformer are quantified into the equivalent number of short-circuit impacts. The comprehensive capability index of the transformer is calculated based on the equivalent number of short-circuit impacts. A long-term degradation prediction model is constructed. The long-term degradation prediction model is used to evaluate the long-term evolution process to obtain the long-term degradation prediction results and generate a long-term degradation prediction report. The transient prediction results are coupled with the long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index. The overall risk status of the transformer is then classified according to the comprehensive risk quantification index.

2. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 1, characterized in that, The process of constructing a four-layer standardized parameter database for the transformer and processing the four-layer standardized parameter database to construct a digital twin model of the transformer includes: Collect the basic parameters, material parameters, structural parameters, and operating parameters of the transformer, and use the basic parameters, material parameters, structural parameters, and operating parameters to construct the four-layer standardized parameter database of the transformer; The four-layer standardized parameter database is verified by performing spatiotemporal consistency verification and physical rule verification on all data. Then, the digital twin model of the transformer is constructed by weighted fusion of multi-source data of the same parameter in the verified four-layer standardized parameter database using a data fusion algorithm.

3. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 1, characterized in that, The process of acquiring the actual short-circuit current waveform of the transformer and extracting short-circuit current features from the actual short-circuit current waveform includes: Extract the actual short-circuit current waveform from the relay protection fault recorder in the substation where the transformer is located, preset the cutoff frequency and use the cutoff frequency to filter the actual short-circuit current waveform, identify the saturation start point of the actual short-circuit current waveform based on the distortion of the actual short-circuit current waveform, and reconstruct the actual short-circuit current waveform during the saturation period using the fitting extrapolation method. The short-circuit current characteristics of the reconstructed real short-circuit current waveform are extracted. The short-circuit current characteristics include the first large half-wave peak current, asymmetry, effective duration, and main frequency components. The effective duration is the period when the amplitude of the real short-circuit current in the real short-circuit current waveform reaches 80% of its peak value. The main frequency components are the main harmonic frequency components other than the power frequency.

4. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 3, characterized in that, The step of inputting short-circuit current excitation into the digital twin model to simulate the transient electromagnetic force, dynamic deformation, and stress distribution of the transformer under short-circuit impact to obtain transient prediction results includes: The three-phase short-circuit current time history curve of the relay protection fault recorder is extracted as the short-circuit current excitation, and the short-circuit current excitation is input into the digital twin model; The transient electromagnetic force of the winding in the transformer under the short-circuit current excitation is calculated using the sequential coupling solution method. The dynamic deformation and stress distribution of the winding are then calculated using the transient electromagnetic force to complete the simulation. Based on all data from the simulation process, the key mechanical indicators of the winding are extracted. The winding is evaluated for deformation or instability risk by comparing the key mechanical indicators with the yield criterion and the instability criterion. The short-time risk index under a single short-circuit impact is calculated. The key mechanical indicators include the maximum equivalent stress, the maximum radial and axial deformation, and the stress concentration factor. The transient prediction result is obtained by summarizing the key mechanical indicators of transient electromagnetic force, dynamic deformation and stress distribution, and short-term risk index.

5. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 2, characterized in that, The construction of a multi-axis degradation model to quantify the long-term evolution of the transformer includes: By aligning and fusing the operating parameters in the four-layer standardized parameter database with multi-source time-series data, a time-series database of equipment operating conditions is established, and key operating characteristic parameters are extracted from the time-series database of equipment operating conditions. The multi-source time-series data includes load rate, oil temperature, vibration, and oil chromatography. The multi-axis degradation model is constructed based on the key operating characteristic parameters. The multi-axis degradation model includes a winding clamping force attenuation model based on the pad creep equation and temperature effect for the mechanical axis, an insulation paper polymerization degree attenuation model and insulation strength correlation model based on the Arrhenius equation for the electrical axis, and a conductor material softening and fatigue characteristic correction model based on cumulative thermal history for the thermal axis. The long-term evolution process of the transformer is quantified through the multi-axis degradation model.

6. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 4, characterized in that, The process of quantifying the transformer's normal operating conditions into equivalent short-circuit impact counts and calculating the transformer's comprehensive capability index based on these equivalent short-circuit impact counts includes: The mechanical stress types and magnitudes generated by various conventional operating conditions on the winding are analyzed, and a conversion algorithm from the conventional operating conditions to the short-circuit impact is established. Based on the conversion algorithm, various conventional operating conditions are quantified into the equivalent number of short-circuit impacts. The total cumulative damage is calculated based on the actual number of short circuits and the equivalent number of short circuit impacts using Miner's linear cumulative damage theory. The key state parameters of the winding are dynamically updated based on the total cumulative damage. The key state parameters include the remaining fatigue life, the current yield strength, and the current clamping force. Using the updated key state parameters, the transformer's short-circuit withstand capability is decomposed into three dimensions: resistance to plastic deformation, resistance to instability, and resistance to fatigue fracture. A weighted comprehensive evaluation method is then used to calculate the comprehensive capability index used to quantify the current remaining short-circuit withstand capability.

7. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 6, characterized in that, The process of constructing a long-term degradation prediction model, using the long-term degradation prediction model to evaluate the long-term evolution process to obtain long-term degradation prediction results, and generating a long-term degradation prediction report includes: A long-term degradation prediction model is constructed by combining time series analysis and physical model extrapolation. The long-term degradation prediction model is then used to make dynamic predictions of clamping force attenuation, insulation aging and cumulative damage growth under different future operating scenarios to obtain the long-term degradation prediction results. The long-term degradation prediction report is generated by combining the comprehensive capability index and the long-term degradation prediction results.

8. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 1, characterized in that, The process involves coupling the transient prediction results with the long-term degradation prediction results to construct a four-quadrant risk assessment matrix and calculating a comprehensive risk quantification index. Based on this comprehensive risk quantification index, the overall risk status of the transformer is classified, including: A two-dimensional risk assessment dataset is constructed based on the transient prediction results and the long-term degradation prediction results. The four-quadrant risk assessment matrix is ​​constructed with the transient prediction results as the horizontal axis and the long-term degradation prediction results as the vertical axis, dividing the dataset into four decision regions: immediate intervention, preventive reinforcement, enhanced monitoring, and routine operation and maintenance, so as to achieve intuitive positioning of risk status. The transient prediction result and the long-term degradation prediction result are coupled by a weighted fusion algorithm to calculate the comprehensive risk quantification index that reflects the overall risk status of the transformer. The comprehensive risk quantification index is then classified into corresponding risk levels according to a preset risk threshold.

9. The method for dynamic verification of transformer short-circuit withstand capability based on transient-time-varying conditions according to claim 8, characterized in that, The step of coupling the transient prediction result with the long-term degradation prediction result to construct a four-quadrant risk assessment matrix and calculate a comprehensive risk quantification index, and classifying the overall risk status of the transformer according to the comprehensive risk quantification index, further includes: The comprehensive risk quantification index of the transformer is mapped to the four-quadrant risk assessment matrix and its corresponding risk area is determined. Based on the comprehensive risk quantification index, the risk level, and the risk area, a comprehensive risk assessment result containing clear risk positioning and classification is generated.

10. A dynamic verification system for the short-circuit withstand capability of a transformer based on transient-time-varying conditions, wherein the verification system applies the verification method according to any one of claims 1 to 9, and the verification system comprises: A digital twin model construction unit is used to construct the four-layer standardized parameter database of the transformer and to process the four-layer standardized parameter database to construct the digital twin model of the transformer. A transient short-circuit impact simulation and prediction unit is connected to the digital twin model construction unit. The transient short-circuit impact simulation and prediction unit is used to acquire the real short-circuit current waveform and extract the short-circuit current characteristics. The short-circuit current characteristics are input into the digital twin model as the short-circuit current excitation to simulate the transient electromagnetic force, dynamic deformation and stress distribution of the transformer under the short-circuit impact to obtain the transient prediction results. Based on the prediction results, a corresponding transient prediction report is generated. A long-term performance degradation modeling and prediction unit is connected to the digital twin model construction unit. The long-term performance degradation modeling and prediction unit is used to construct the multi-axis degradation model to quantify the long-term evolution process of the transformer, quantify the normal operating conditions into the equivalent short-circuit impact number, calculate the comprehensive capability index based on the equivalent short-circuit impact number, construct the long-term degradation prediction model, use the long-term degradation prediction model to evaluate the long-term evolution process to obtain the long-term degradation prediction result, and generate the long-term degradation prediction report. The comprehensive risk assessment unit is connected to the long-term performance degradation modeling and prediction unit and the transient short-circuit impact simulation and prediction unit. The comprehensive risk assessment unit is used to couple the transient prediction results with the long-term degradation prediction results, construct the four-quadrant risk assessment matrix and calculate the comprehensive risk quantification index, and classify the overall risk status of the transformer according to the comprehensive risk quantification index.

Images