Lightning impulse experiment data judgment model construction system based on digital twinning

Abstract

The invention discloses a lightning impulse experiment data judgment model construction system based on digital twinning, and the system comprises a digital twinning rehearsal unit which is used for analyzing a multi-physics field coupling model according to a finite element; the quantum-level waveform synthesis module is used for combining a waveform jitter suppression algorithm of a quantum random number generator, generating pulses by using the received parameters, and forming a waveform as a signal source basis; the physical generation feedback correction unit is used for driving a physical waveform generator to output a waveform; the self-adaptive phase locking feedback module is used for collecting target waveform and actual output waveform data in real time; the method has the advantages that digital twinborn rehearsal, physical generation and feedback correction closed loop are established, bidirectional optimization of a physical world and a digital model is achieved through finite element analysis and the quantum electric field sensor, the nonstandard waveform reduction degree is improved, and the real working condition in the complex environment is effectively simulated.

Description

Technical Field

[0001] This invention relates to the field of experimental data systems, specifically to a system for constructing a judgment model for lightning impact experimental data based on digital twins. Background Technology

[0002] Traditional experimental systems face the dual challenges of scenario limitations and data lag. For example, in the field of lightning protection testing for power equipment, due to physical experimental conditions, traditional methods can only cover less than 1% of extreme operating conditions. Simulation models based on historical data cannot dynamically adapt to complex environmental changes, resulting in a missed detection rate of over 30% for new fault modes. With the increasing frequency of extreme weather events globally, such as a 27% increase in lightning activity intensity over the past decade, and the exponential growth in reliability requirements for new power equipment such as third-generation semiconductor devices and space-based equipment, traditional experimental models can no longer meet the needs of technological innovation. At the same time, breakthroughs in quantum sensing, deep learning, and distributed computing technologies have made it possible to build more intelligent and efficient experimental systems. Digital twin technology, with its virtual-real mapping capabilities, has become a core path to break through the bottlenecks of traditional experiments, driving the industry's transformation from experience-driven to data-driven. Against this backdrop, a technological closed loop of waveform control, parallel deduction, and twin evolution has emerged, aiming to achieve a fundamental transformation of the experimental paradigm through technological integration. Summary of the Invention

[0003] The purpose of this invention is to solve the problems mentioned above, and therefore proposes a system for constructing a judgment model for lightning impact experimental data based on digital twins.

[0004] The objective of this invention can be achieved through the following technical solution: a system for constructing a judgment model based on lightning impact experimental data using digital twins, comprising:

[0005] The digital twin pre-simulation unit is used to simulate the impact of the target waveform on the equipment under different environmental conditions based on finite element analysis and multiphysics coupling model. It predicts the parameters required to generate the target waveform through Monte Carlo simulation method and transmits the parameters to the quantum-level waveform synthesis module and the physical generation feedback correction unit at the same time.

[0006] The quantum-level waveform synthesis module is used to combine the waveform jitter suppression algorithm of the quantum random number generator to generate pulses using the received parameters, forming a waveform as the basis of the signal source;

[0007] The physical generation feedback correction unit is used to drive the physical waveform generator to output waveforms and uses a quantum electric field sensor to collect waveform data in real time. The collected data is compared with the target waveform of the digital twin pre-simulation unit, and the digital model parameters are corrected through the back propagation algorithm to form closed-loop control.

[0008] The adaptive phase-locked feedback module is used to acquire target waveform and actual output waveform data in real time. It uses a waveform tracking neural network based on deep learning to extract features and perform comparative analysis to determine whether there is amplitude attenuation or phase shift anomaly.

[0009] Furthermore, the digital twin pre-simulation unit, together with the quantum-level waveform synthesis module, the physical generation feedback correction unit, and the adaptive phase-locked feedback module, constitutes a waveform control unit, which is used for waveform restoration processing.

[0010] Furthermore, the waveform restoration process is as follows:

[0011] S1: Predict the parameters required to generate the target waveform through the digital twin pre-simulation unit, and transmit these parameters simultaneously to the quantum-level waveform synthesis module and the physical generation feedback correction unit;

[0012] S2: The quantum-level waveform synthesis module combines the waveform jitter suppression algorithm of the quantum random number generator to generate pulses using the received parameters, forming a waveform as the basis of the signal source;

[0013] S3: The physical generation feedback correction unit drives the physical waveform generator to output the initial physical waveform based on the parameters provided by the digital twin pre-simulation unit;

[0014] S4: The physical generation feedback correction unit uses a quantum electric field sensor to collect waveform data in real time, compares it with the results of the digital twin pre-simulation, and simultaneously synchronizes the real-time state of the waveform to the adaptive phase-locked feedback module. When the physical generation feedback correction unit finds that there is a deviation between the collected waveform data and the parameters generated by the digital twin pre-simulation unit, it corrects the digital model parameters through the backpropagation algorithm and feeds back the optimized parameters to the digital twin pre-simulation unit.

[0015] S5: The digital twin pre-simulation unit re-evaluates the waveform generation strategy based on the new parameters, and then sends the adjusted instructions to the quantum-level waveform synthesis module and the adaptive phase-locked feedback module.

[0016] S6: The quantum-level waveform synthesis module adjusts the voltage source output according to the new parameters, and the adaptive phase-locked feedback module automatically adjusts the capacitance and inductance parameters of the pulse generator through the FPGA.

[0017] Furthermore, the system also includes a scene parametric modeling module for constructing a digital experimental universe, the specific process of which is as follows:

[0018] A1: Obtain non-standard waveform data from the quantum-level waveform synthesis module and the adaptive phase-locked feedback module, and use it as the basis for waveform parameters in the scene gene fragment;

[0019] A2: Use the parameters generated by the digital twin pre-simulation unit during the simulation process as the core basis for lightning parameters and equipment response parameters in the scene gene;

[0020] A3: Based on real-time monitoring data from the adaptive phase-locked feedback module, the scene gene fragments are dynamically corrected.

[0021] Furthermore, the system also includes an automatic exploration module for counterintuitive scenarios, which introduces a curiosity-driven mechanism based on deep reinforcement learning to construct a reward function to guide the active exploration of unconventional scenarios in the digital experimental universe. When the system detects that the combination of environmental parameters deviates from the cognitive range, it triggers exploration behavior.

[0022] Furthermore, the system also includes a cross-temporal fault chain inference module, which uses a spatiotemporal folding algorithm to mathematically model the physical and chemical changes during the equipment aging process. Through numerical calculation methods, it simulates the lightning impact cycle of the transformer, and can completely infer the entire life cycle evolution process of its insulation material from crack initiation to breakdown, thereby locating the critical point of cumulative effect.

[0023] Furthermore, the automatic exploration module for counterintuitive scenarios is also used for verification processing of extreme working conditions, climate change, and space environment.

[0024] Furthermore, the verification process for the extreme operating conditions is as follows:

[0025] Based on the finite element analysis and multiphysics coupling model of the digital twin pre-simulation unit, a digital twin of the device under unconventional operating conditions is constructed.

[0026] The spatiotemporal folding algorithm of the cross-spatiotemporal fault chain inference module is used to compress and simulate the device aging process;

[0027] By leveraging the deep reinforcement learning mechanism of the counterintuitive scenario automatic exploration module, combined with a preset failure threshold, it proactively identifies deviations from conventional performance parameter changes, ultimately uncovering potential failure modes that traditional experiments cannot cover.

[0028] Furthermore, the verification process for the aforementioned climate change is as follows:

[0029] A digital experimental universe containing different climate zones was constructed using the scene parametric modeling module;

[0030] By introducing a curiosity-driven mechanism based on deep reinforcement learning, a reward function is constructed to guide active exploration of unconventional scenarios. Potential lightning phenomena and risks are explored under different climatic conditions. Using a spatiotemporal folding algorithm, the physical and chemical changes of equipment during the aging process under different climatic conditions are mathematically modeled. Through numerical calculation methods, the entire life cycle evolution process of equipment under the influence of long-term climate change is simulated, the critical point of cumulative effect is located, and the future weather is predicted by combining digital experimental universe. The lightning strike patterns under different climatic conditions are simulated, and the lightning failure rate is statistically analyzed.

[0031] Furthermore, the verification process for the aforementioned space environment is as follows:

[0032] The experimental scenario is constructed by calling the relevant waveform and parameter data of plasma lightning through the scenario parameterization modeling module. The quantum-level waveform synthesis module, adaptive phase-locked feedback module and physical generation feedback correction unit work together to generate a waveform containing particle bombardment through waveform control technology. The waveform is then applied to the digital twin pre-simulation unit for simulation verification to test the effectiveness of the protection design.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] Quantum-level waveform synthesis and dynamic locking technology enables quantum-level precise control of lightning waveforms, allowing for waveform reproduction and improved waveform stability, thus providing a more realistic and reliable signal source for equipment testing.

[0035] Establish a closed loop of digital twin pre-simulation, physical generation, and feedback correction. Through finite element analysis and quantum electric field sensors, achieve bidirectional optimization of the physical world and digital model to improve the fidelity of non-standard waveforms and effectively simulate real working conditions in complex environments.

[0036] By combining scenario gene coding technology with genetic algorithms, a hybrid computing cluster of quantum front-end and back-end was constructed, increasing the experimental scenario generation capability from 10 3 Level upgraded to 10 7 The computing efficiency is significantly improved, greatly expanding the scope of exploration in extreme scenarios;

[0037] Autonomous scenario exploration and fault simulation introduce a curiosity-driven reinforcement learning mechanism and use a spatiotemporal folding algorithm to realize cross-spatiotemporal fault chain simulation to actively discover equipment failure modes, accurately locate the critical point of cumulative effect in the equipment aging process, apply waveform control and parallel universe technology to power equipment testing, climate change lightning protection optimization and space environment, simulate new equipment testing cycle shortens and costs are reduced, promote the transformation of lightning protection standards to data-driven, and reduce equipment lightning strike accident rate. Attached Figure Description

[0038] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0039] Figure 1 This is a schematic diagram of a system for constructing a judgment model for lightning impact experimental data based on digital twins, according to the present invention. Detailed Implementation

[0040] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] In existing technologies, traditional experimental systems are limited by physical conditions and cannot cover complex and ever-changing extreme working conditions. The field of lightning protection testing for power equipment has long faced problems such as single experimental scenarios and lagging data updates. Simulation models built from historical data cannot dynamically adapt to changes in environmental parameters, resulting in significant defects in the identification of new fault modes. With the development of quantum sensing and distributed computing technologies, intelligent upgrades of experimental systems have become possible, but existing technologies have not yet achieved closed-loop control for real-time interaction and dynamic correction of virtual and real data.

[0042] To address the aforementioned issues, researchers observed that traditional waveform generation systems suffer from a disconnect between static parameter settings and the physical environment. Through reverse analysis of multiple lightning strike incidents, they discovered a nonlinear correlation between waveform distortion and equipment failure. By studying the stability characteristics of quantum voltage sources, they attempted to integrate digital twin simulation with the physical generation system for real-time data interaction. Furthermore, they combined deep learning algorithms to dynamically track waveform features, ultimately forming a closed-loop control architecture that links the virtual and real worlds.

[0043] For this, please refer to Figure 1 As shown, this application proposes a lightning impact experiment data judgment model construction system based on digital twins, including a digital twin pre-simulation unit, a quantum-level waveform synthesis module, a physical generation feedback correction unit, and an adaptive phase-locked feedback module. The digital twin pre-simulation unit performs environmental simulation through finite element analysis and a multi-physics coupling model. The quantum-level waveform synthesis module uses a Josephson junction array voltage source to generate ultrashort pulses. The physical generation feedback correction unit realizes data acquisition and model correction through a quantum electric field sensor. The adaptive phase-locked feedback module uses a deep learning neural network to perform waveform feature analysis.

[0044] Among them, the digital twin pre-simulation unit refers to a virtual simulation platform based on multi-physics coupling, which can use ANSYS Maxwell software to realize the joint simulation of electromagnetic and thermal fields, and is used to predict the changes of waveform parameters under different environmental conditions. The quantum-level waveform synthesis module refers to a waveform generator based on quantum voltage reference, which can generate stable pulse signals through Josephson junction array in low-temperature superconducting environment. The physical generation feedback correction unit refers to a closed-loop control system that includes quantum sensing and algorithm correction, which can use diamond nitrogen-vacancy color center sensor to realize picosecond-level electric field measurement. The adaptive phase-locked feedback module refers to a waveform tracking system with dynamic adjustment capability, which can use convolutional neural network to extract waveform time-frequency features.

[0045] Specifically, the finite element model first simulates the electromagnetic field distribution of the target device under lightning strike, generating an initial waveform parameter set. The Josephson junction array outputs a reference pulse signal according to the parameter command, the physical generator synchronously generates the actual waveform, and the quantum electric field sensor captures waveform distortion data in real time. By comparing the difference between the simulation prediction value and the measured value, the backpropagation algorithm automatically adjusts the boundary conditions of the finite element model. The phase locking module continuously monitors the waveform phase shift, and after identifying abnormal patterns through the neural network, it triggers parameter correction commands, forming a closed-loop control chain from virtual prediction to physical generation and then to data feedback.

[0046] Compared with existing technologies, traditional systems adopt an open-loop control mode, and the waveform parameters cannot be dynamically adjusted according to environmental changes after being set. This solution achieves millisecond-level feedback adjustment through quantum sensing and deep learning technologies, solving the problem of mismatch between physical environment and digital model. Existing technologies rely on a single physical generator to output waveforms. This solution significantly improves waveform fidelity under complex working conditions through the collaborative work of digital twin pre-simulation and quantum synthesis modules.

[0047] Through the above technical solutions, this application achieves accurate simulation and dynamic optimization of the lightning impulse test environment. The waveform generation system can automatically correct signal distortion caused by changes in environmental parameters, effectively identify transient anomalies that are difficult to capture by traditional methods, and the virtual-real data interaction mechanism ensures the real-time consistency between the experimental model and the actual working conditions, providing a high-precision test platform for the reliability assessment of power equipment.

[0048] This application further proposes a waveform control unit consisting of a digital twin pre-simulation unit, a quantum-level waveform synthesis module, a physical generation feedback correction unit, and an adaptive phase-locked feedback module. This unit is used for waveform restoration processing.

[0049] Among them, the digital twin pre-simulation unit refers to the module that performs simulation based on finite element analysis and multiphysics coupling model. Specifically, it can use Monte Carlo simulation method to predict target waveform parameters. Its role is to provide dynamic parameter benchmark for waveform generation. The quantum-level waveform synthesis module refers to the hardware unit that combines Josephson junction array voltage source and quantum random number generator. Specifically, it realizes high-precision signal source through picosecond-level ultrashort pulse generation technology. Its role is to transform simulation parameters into physically realizable waveform basis. The physical generation feedback correction unit refers to the closed-loop control module that collects data through quantum electric field sensor and compares it with simulation results. Specifically, it uses backpropagation algorithm to correct model parameters. Its role is to eliminate deviations in the physical generation process. The adaptive phase-locked feedback module refers to the real-time monitoring unit that tracks waveforms based on deep learning. Specifically, it judges waveform anomalies through feature extraction and comparative analysis. Its role is to maintain the stability of output waveform.

[0050] Specifically, the waveform control unit achieves waveform restoration through multi-module collaboration. The parameters generated by the digital twin pre-simulation unit are synchronously transmitted to the quantum-level waveform synthesis module and the physical generation feedback correction unit. The former generates a high-precision initial waveform, and the latter drives the physical generator to output the actual waveform. After the quantum electric field sensor collects the physical waveform data in real time, it compares it with the target parameters of the pre-simulation unit. When an amplitude or phase deviation is detected, the model parameters are corrected through the backpropagation algorithm and fed back to the pre-simulation unit. The adaptive phase-locked feedback module continuously monitors the waveform status, uses neural networks to analyze abnormal features, and sends adjustment commands to the synthesis module and the feedback correction unit to form a dynamic closed-loop control.

[0051] Compared with existing technologies, traditional waveform restoration systems rely on single parameter generation and static feedback mechanisms, which cannot dynamically adapt to waveform distortion caused by environmental changes. This solution achieves dynamic optimization and closed-loop correction of waveform parameters by combining digital twin pre-simulation and quantum-level synthesis with real-time data acquisition and deep learning analysis, thus solving the problem of high waveform distortion rate in traditional systems under complex environments.

[0052] Through the above technical solution, this application achieves high-fidelity waveform restoration capability, and can automatically adjust the generation strategy when environmental parameters fluctuate, ensuring the consistency between the physical output waveform and the simulation target, and providing a precise and controllable test signal source for lightning impact experiments.

[0053] This application further proposes a waveform restoration processing method for a lightning impact experimental data judgment model construction system based on digital twins, including: predicting the parameters required to generate the target waveform through a digital twin pre-simulation unit, and simultaneously transmitting these parameters to a quantum-level waveform synthesis module and a physical generation feedback correction unit; the quantum-level waveform synthesis module generates ultrashort pulses with picosecond-level rising edges by combining a waveform jitter suppression algorithm of a quantum random number generator; the physical generation feedback correction unit drives a physical waveform generator to output an initial physical waveform, collects waveform data in real time using a quantum electric field sensor and compares it with the digital twin pre-simulation results, and corrects the digital model parameters through a backpropagation algorithm; the digital twin pre-simulation unit re-evaluates the waveform generation strategy based on the new parameters and sends adjustment commands; the quantum-level waveform synthesis module adjusts the output of the Josephson junction array voltage source, and the adaptive phase-locked feedback module automatically adjusts the capacitance and inductance parameters of the pulse generator through an FPGA;

[0054] Among them, the digital twin pre-simulation unit refers to the simulation module based on finite element analysis and multiphysics coupling model, which can be implemented using ANSYS or COMSOL software, and is used to predict waveform parameters under different environmental conditions; the quantum-level waveform synthesis module refers to the signal generation device using Josephson junction array voltage source, which can be implemented using superconducting quantum interference device, and is used to generate high-precision ultrashort pulses; the physical generation feedback correction unit refers to the closed-loop control system containing quantum electric field sensor, which can be implemented using diamond nitrogen-vacancy color center sensor, and is used to collect waveform data in real time and correct model parameters; the backpropagation algorithm refers to the neural network training method based on error gradient descent, which can be implemented using TensorFlow framework, and is used to optimize digital model parameters; the FPGA automatic adjustment refers to the fast parameter adjustment mechanism based on hardware description language, which can be implemented using Xilinx chip, and is used to realize real-time correction of capacitance and inductance parameters;

[0055] Specifically, the waveform restoration process generates an initial parameter set, such as the pulse frequency range and amplitude threshold range, through a digital twin pre-simulation unit. These parameters are then synchronously transmitted to the quantum-level waveform synthesis module and the physical generation feedback correction unit. The quantum-level waveform synthesis module uses a Josephson junction array to generate a reference waveform signal and eliminates signal jitter interference through a quantum random number generator, forming an initial waveform with picosecond-level precision. When the physical generation feedback correction unit drives the physical waveform generator to output the actual waveform, it continuously collects quantum electric field sensor data and compares it with the pre-simulation model. When the amplitude deviation exceeds a preset threshold, it triggers a backpropagation algorithm to iteratively update the model weight matrix. The updated parameters are then re-input into the digital twin pre-simulation unit for multi-physics coupling calculations to generate optimized waveform control commands. Finally, the FPGA hardware circuit performs dynamic parameter matching on the energy storage element of the pulse generator to complete the closed-loop optimization of the waveform parameters.

[0056] Compared with existing technologies, traditional waveform restoration methods rely on a single physical generator for open-loop control, which cannot correct waveform distortion caused by environmental interference in real time. This solution achieves dynamic optimization of waveform parameters by constructing a closed-loop feedback mechanism between a digital twin model and a physical device, supported by quantum sensing technology. In existing technologies, waveform adjustment response time usually takes minutes. This solution uses FPGA hardware acceleration technology to improve the feedback adjustment speed to the nanosecond level. Traditional systems using fixed parameter models are difficult to adapt to complex environmental changes. This solution achieves online self-learning of model parameters through backpropagation algorithm.

[0057] Through the above technical solutions, this application effectively solves the problem of insufficient output accuracy of physical waveform generators. By combining quantum-level signal synthesis with real-time feedback correction, it ensures the consistency between the state of the digital model and the physical device. The closed-loop control mechanism can automatically compensate for waveform distortion caused by environmental interference, improve the reliability of lightning impact test data, and the application of hardware acceleration technology significantly shortens the parameter adjustment cycle, enabling the waveform restoration process to dynamically adapt to complex working conditions.

[0058] This application further proposes a system including a scene parameterized modeling module, which is used to construct a digital experimental universe. The specific process includes: obtaining non-standard waveform data from a quantum-level waveform synthesis module and an adaptive phase-locked feedback module as the waveform parameter basis for scene gene fragments; using parameters generated by the digital twin pre-simulation unit during the simulation process as the core basis for lightning parameters and equipment response parameters in the scene gene; dynamically correcting the scene gene fragments based on real-time monitoring data from the adaptive phase-locked feedback module; generating experimental scenarios by referring to historical valid data when performing crossover and mutation operations on gene fragments using a genetic algorithm; and adopting a quantum front-end plus classical back-end architecture, where the quantum computer uses the principle of quantum superposition to process the combination calculation of scene genes in parallel, and the classical server cluster uses containerization technology for dynamic resource scheduling and is responsible for multi-physics simulation calculations.

[0059] Among them, the scene parameterization modeling module refers to the basic architecture used to build a digital experimental environment. Specifically, it can be implemented using multi-source data fusion and dynamic parameter adjustment technology. It constructs a virtual experimental scene by integrating waveform parameters, lightning parameters, and equipment response parameters. The scene gene fragment refers to the encoding unit containing the core parameters of the experimental scene. Specifically, it can be implemented using multi-dimensional data matrices and genetic algorithm operation objects. It is used to store and transmit key information such as waveform characteristics, environmental conditions, and equipment status. Dynamic correction refers to adjusting scene parameters based on real-time monitoring data. Specifically, it can be implemented using feedback control algorithms and adaptive learning models to ensure that the experimental scene remains synchronized with the physical world. The quantum front-end plus classical back-end architecture refers to a hybrid computing framework that combines quantum computing and classical computing. Specifically, it can be implemented using quantum processors for parallel computing task allocation and classical server clusters for physical field simulation calculations. It accelerates the generation of large-scale scene combinations through the principle of quantum state superposition.

[0060] Specifically, the scene parameterization modeling module integrates non-standard waveform data from the quantum-level waveform synthesis module with simulation parameters from the digital twin pre-simulation unit to construct a multi-dimensional scene gene containing lightning characteristics and equipment status. When the adaptive phase-locked feedback module detects amplitude attenuation in the actual waveform transmission, this module dynamically corrects the situation by adjusting the corresponding parameters in the scene gene. When generating experimental scenarios, a genetic algorithm is used to extract features from historical valid data, and combined experimental scenarios are generated through crossover and mutation operations. The quantum computing part utilizes the superposition characteristics of qubits to process the combined computation tasks of scene genes in parallel, such as simulating multiple waveform combinations simultaneously. The classical computing part dynamically allocates computing resources through containerization technology to complete multi-physics coupling simulations. For example, when simulating lightning characteristics in different latitude regions, distributed computing is used to improve computational efficiency.

[0061] Compared with existing technologies, traditional systems rely on fixed parameter libraries to generate limited experimental scenarios and have low computational resource scheduling efficiency. This solution improves the realism of the scenarios through a dynamic parameter correction mechanism, accelerates the scenario generation process by using quantum parallel computing, and combines containerization technology to achieve elastic allocation of computing resources, thereby significantly improving the computational efficiency of complex environment simulation.

[0062] Through the above technical solutions, this application solves the problem of limited ability to generate traditional experimental scenarios, can automatically construct massive combined experimental scenarios and maintain parameter authenticity, and significantly shortens the simulation time of multiphysics fields through quantum-classical hybrid computing architecture, providing effective support for equipment performance testing under extreme environmental conditions.

[0063] This application further proposes an automatic exploration module for counterintuitive scenarios, which introduces a curiosity-driven mechanism based on deep reinforcement learning, constructs a reward function to guide the active exploration of unconventional scenarios in the digital experimental universe, and triggers exploration behavior when the system detects that the combination of environmental parameters deviates from the conventional cognitive range.

[0064] Among them, the curiosity-driven mechanism of deep reinforcement learning refers to simulating the human drive to explore the unknown by constructing an intrinsic reward function. Specifically, it can be implemented using a reward generation algorithm based on prediction error. This mechanism incentivizes the system to actively explore abnormal regions in the parameter space by quantifying the degree of deviation between the environmental state and the expected model. The reward function-guided active exploration refers to transforming the discovery probability of unconventional scenarios into an optimizable mathematical index. Specifically, it can be implemented using a reward function design method based on information gain. By maximizing the difference in information entropy obtained during the exploration process, it drives the system to prioritize accessing parameter combinations that have not been fully studied. The deviation of environmental parameter combinations from the conventional cognitive range refers to the system identifying parameter combinations that are statistically significantly deviated from the mean range by comparing the typical parameter distributions in the historical experimental database. Specifically, it can be implemented using an anomaly detection algorithm based on Mahalanobis distance. When the distance between the parameter vector and the covariance matrix of the conventional scene exceeds a set threshold, the exploration process is triggered.

[0065] Specifically, during the operation of the digital experimental universe, the counterintuitive scenario automatic exploration module continuously monitors the statistical distribution characteristics of environmental parameter combinations. When it detects that the humidity parameter is below the normal threshold and the lightning frequency is abnormally reduced, the module automatically generates an experimental scheme that includes ultra-low humidity and ultra-low frequency lightning parameters. The corresponding physical waveform is generated through the quantum-level waveform synthesis module, and the long-term operating state of the equipment under this combination of conditions is simulated using the cross-space-time fault chain deduction module. During the simulation, the module continuously collects the breakdown voltage data of the insulating material. When it finds that the breakdown voltage value deviates systematically from the predicted value of the conventional cognitive model, it automatically marks the anomaly as a counterintuitive discovery and updates the parameter prediction model of the digital twin pre-simulation unit.

[0066] Compared with existing technologies, traditional experimental systems rely on manually preset scene parameters, which makes it difficult to break through the experience boundaries of engineers. This solution can effectively discover hidden abnormal correlations in the parameter space by constructing an autonomous exploration mechanism. For example, in ultra-low frequency lightning scenarios, traditional methods are limited by the frequency adjustment range of the physical generator and cannot generate lightning waveforms below 50Hz for testing. However, this system uses the wideband adjustment capability of the quantum-level waveform synthesis module and combines it with an autonomous exploration algorithm to achieve effective verification of this scenario.

[0067] Through the above technical solution, this application can overcome the subjective limitations of traditional experimental design and automatically identify the abnormal response characteristics of equipment under special environmental coupling conditions. For example, in the scenario of ultra-low humidity and low-frequency lightning, the system successfully captured the nonlinear decrease phenomenon of the breakdown voltage of the insulating material. This discovery provides a new theoretical basis for optimizing the protection design of the equipment and significantly improves the scenario coverage capability and anomaly detection reliability of the experimental system.

[0068] This application further proposes a cross-temporal fault chain inference module, which uses a spatiotemporal folding algorithm to mathematically model the physical and chemical changes during the equipment aging process, and compresses the long-term aging process into a short-term simulation through numerical calculation methods. In the lightning impact cycle simulation, it infers the entire life cycle evolution process of the insulating material from micro-crack initiation to macro-breakdown, and accurately locates the critical point of cumulative effect. When the module runs, it calls the pulse frequency, amplitude threshold, and phase offset range generated by the digital twin pre-simulation unit as the excitation basis for the equipment operating conditions, uses non-standard waveform data output by the quantum-level waveform synthesis module and the adaptive phase-locked feedback module to simulate extreme electrical stress scenarios, adjusts the simulation parameters by combining the real-time monitoring data of the adaptive phase-locked feedback module, and uses the tens of millions of experimental scenarios generated by the scene parameterization modeling module as simulation materials. At the same time, the inference results are fed back to the scene gene optimization.

[0069] Among them, the spatiotemporal folding algorithm refers to a calculation method that compresses the time dimension of the equipment aging process through mathematical modeling. Specifically, it can be implemented by combining the numerical solution of partial differential equations with a time acceleration factor. It is used to simulate the long-term aging process within a finite time. The parameters generated by the digital twin pre-simulation unit refer to the equipment operating condition parameters obtained through finite element analysis and multi-physics coupling models. Specifically, they can be implemented by using the pulse characteristic parameters output by electromagnetic field simulation software. They serve as the reference data for the equipment under external excitation. The non-standard waveform data output by the quantum-level waveform synthesis module refers to the lightning impact waveform containing abnormal characteristics. Specifically, it can be implemented by using the distorted waveform data generated by the Josephson junction array voltage source. It is used to construct the equipment operation scenario under extreme electrical stress conditions. The real-time monitoring data of the adaptive phase-locked feedback module refers to the dynamic change information of the actual output waveform. Specifically, it can be implemented by using the amplitude and phase deviation data collected by the quantum electric field sensor. It is used to correct the parameter deviation in the simulation process. The tens of millions of experimental scenarios generated by the scene parameterization modeling module refer to the environmental conditions and equipment status datasets combined by genetic algorithms. Specifically, they can be implemented by using parameter encoding and random combination algorithms. They provide diverse input conditions for fault inference.

[0070] Specifically, in the transformer insulation aging simulation process, the altitude and lightning parameter correspondence provided by the digital twin pre-simulation unit are first called to obtain the optimal waveform parameters of lightning impact at different altitudes. The quantum-level waveform synthesis module generates non-standard waveforms containing double-pulse characteristics based on the parameters to simulate abnormal lightning strikes that may occur in actual operation. The adaptive phase-locked feedback module monitors the phase offset in the waveform transmission process in real time. When a time deviation of 0.5 microseconds is detected, the dynamic adjustment mechanism of the simulation parameters is immediately triggered. The scene parameterization modeling module provides a combined scene including temperature gradient changes and material aging degree. For example, the aging degree of insulation material is correlated with humidity changes. The cross-temporal fault chain inference module compresses the time scale of the material microstructure evolution process by three orders of magnitude through the spatiotemporal folding algorithm, and completes the simulation of the crack propagation process that would require 100 years of observation by traditional methods in 1 hour. At the same time, combined with the real-time adjusted waveform parameters and equipment status parameters, the critical electric field strength threshold for insulation breakdown is accurately calculated.

[0071] Compared with existing technologies, traditional equipment aging simulation methods are limited by the time span of physical experiments and can only obtain limited data points through accelerated aging tests. They cannot fully present the continuous evolution process of material failure. In existing technologies, the correlation analysis between waveform parameters and equipment state parameters uses a static database, which is difficult to respond to dynamic changes in actual operation in real time. This solution realizes the compressed simulation of long-term processes through a spatiotemporal folding algorithm. Combined with a real-time data feedback mechanism, it can dynamically correct simulation parameters and capture the impact of transient anomalies on equipment life.

[0072] Through the above technical solutions, this application solves the problem of incomplete experimental data caused by the excessively long aging simulation cycle of traditional equipment, overcomes the defect that static parameter databases cannot adapt to the dynamic changes of actual working conditions, realizes the accurate prediction of the failure process of the entire life cycle of equipment under extreme electrical stress conditions, and ensures that the simulation process is synchronized with the actual operating state by establishing a multi-module data linkage mechanism, effectively identifying the critical point of cumulative damage of insulating materials in complex environments.

[0073] This application further proposes that the counterintuitive scenario automatic exploration module can also be used for verification processing of extreme working conditions, climate change, and space environment.

[0074] The verification process for extreme operating conditions involves generating various combinations of extreme environmental conditions using parallel universe technology, such as scenarios with superimposed high temperatures and high currents. This is combined with waveform control technology to simulate corresponding waveforms and discover potential failure modes of devices under unconventional conditions. Specifically, a digital twin pre-simulation unit can be used to construct a digital twin of the device, and the spatiotemporal folding algorithm of the cross-spatiotemporal fault chain inference module can be used to compress the aging process simulation. Deep reinforcement learning mechanisms can be used to identify deviations in performance parameter changes. The verification process for climate change involves constructing a digital experimental universe containing different climate zones, combining climate models to predict future extreme weather trends, and simulating lightning strike patterns under different climate conditions. Specifically, a scene parameterization modeling module can be used to generate climate zone parameters, and a deep reinforcement learning mechanism can be used to actively explore special lightning phenomena. A spatiotemporal folding algorithm can be used to simulate the evolution of equipment under long-term climate change. The verification process for the space environment involves calling plasma lightning waveform and parameter data to construct experimental scenarios, generating waveforms containing particle bombardment for protective design testing. Specifically, a quantum-level waveform synthesis module can be used to generate specific waveforms, and an adaptive phase-locked feedback module and a physical generation feedback correction unit can be used to adjust waveform parameters. The digital twin pre-simulation unit can then be used for simulation verification.

[0075] Specifically, in extreme operating condition verification, parallel universe technology generates extreme conditions such as the combination of high temperature and high frequency oscillating lightning; the digital twin pre-simulation unit constructs a digital twin of the device; the cross-temporal fault chain inference module compresses the aging process simulation; the counterintuitive scenario automatic exploration module identifies abnormal parameter changes through preset failure thresholds, and finally discovers failure modes not covered by traditional experiments. In climate change verification, the scenario parameterization modeling module integrates climate zone parameters and future weather forecast data; the deep reinforcement learning mechanism actively explores special lightning phenomena in environments such as polar and desert environments; the spatiotemporal folding algorithm simulates the aging process of equipment under long-term climate change and statistically analyzes the trend of lightning failure rate changes. In space environment verification, the quantum-level waveform synthesis module generates waveforms containing particle bombardment based on plasma lightning data; the adaptive phase-locked feedback module adjusts waveform parameters in real time; the physical generation feedback correction unit drives the physical generator to output waveforms; and the digital twin pre-simulation unit verifies the effectiveness of the protection design.

[0076] Compared with existing technologies, traditional experimental systems are limited by physical conditions and can only cover normal working conditions. They cannot simulate extreme environmental combinations or the impact of long-term climate change. Furthermore, ground-based experiments are difficult to reproduce space scenarios. This solution generates non-standard waveforms through parallel universe technology and quantum-level waveform synthesis modules. Combined with digital twin pre-simulation and spatiotemporal folding algorithms, it achieves multi-dimensional scene coverage, solving the problem of limited testing range of traditional methods.

[0077] Through the above technical solutions, this application can improve the coverage of extreme working condition tests, identify potential failure risks of devices in unconventional environments in advance, accurately predict the impact of climate change on lightning strike failure rate, optimize equipment protection strategies, effectively verify protection design in extreme space scenarios, and fill the gap in ground-based experimental capabilities.

[0078] This application further proposes a verification and processing method for extreme working conditions. It generates various extreme working conditions through parallel universe technology, accurately simulates the corresponding waveforms using waveform control technology, analyzes the potential failure modes of devices under combined working conditions through digital twin models, constructs a digital twin of the device under unconventional working conditions based on finite element analysis and multiphysics coupling model of digital twin pre-simulation unit, compresses the device aging simulation process using the spatiotemporal folding algorithm of cross-spatiotemporal fault chain inference module, and actively identifies deviations from the norm in performance parameter changes through the deep reinforcement learning mechanism of the counterintuitive scenario automatic exploration module combined with failure threshold.

[0079] Among them, parallel universe technology refers to the technology of generating multiple independent experimental scenarios in parallel through distributed computing resources. Specifically, it can be implemented using a hybrid architecture of quantum computing and classical computing. It utilizes the principle of quantum superposition to achieve the parallel generation of large-scale scenarios, solving the problem of insufficient coverage of traditional experimental scenarios. Waveform control technology refers to the technology of coordinating the adjustment of the output waveform through a quantum-level waveform synthesis module and an adaptive phase-locked feedback module. Specifically, it uses a Josephson junction array voltage source combined with a quantum random number suppression algorithm to ensure accurate reproduction of waveforms under extreme conditions. Spacetime folding algorithm refers to the technology of mathematically modeling and accelerating the simulation of the physical and chemical changes in the aging process of equipment. Specifically, it uses numerical calculation methods to quickly extrapolate long-term effects by compressing the time dimension. Deep reinforcement learning mechanism refers to the technology of autonomously identifying abnormal parameter changes based on a preset failure threshold. Specifically, it uses a neural network model to construct a reward function to drive exploration behavior and actively capture potential failure modes that are difficult to detect in traditional experiments.

[0080] Specifically, in the extreme operating condition verification process, the parallel universe technology is first used to generate a combination of operating conditions including high temperature, high current and steep voltage change rate, such as a combination of high temperature and high frequency oscillating lightning. Then, the waveform control technology generates a precisely matched physical waveform based on the parameters provided by the digital twin pre-simulation unit. The quantum-level waveform synthesis module outputs picosecond-level pulses through the Josephson junction array. The adaptive phase-locked feedback module adjusts the capacitance and inductance parameters in real time to maintain waveform stability. The digital twin model synchronously constructs a multi-physics field coupled simulation of the device. The cross-temporal fault chain inference module compresses the device aging process to a calculable time scale. The counterintuitive scenario automatic exploration module continuously monitors parameters such as temperature gradient and electric field distribution. When a local electric field distortion exceeds the preset threshold, the failure mode analysis process is automatically triggered. The whole process forms a closed-loop verification system, and the potential risks under extreme operating conditions are proactively discovered through the collaboration of multiple modules.

[0081] Compared with existing technologies, traditional experimental methods are limited by physical conditions and can only test a limited number of conventional working conditions. They also rely on manually setting experimental parameters and cannot effectively capture abnormal failures caused by complex combinations of working conditions. This solution uses parallel universe technology to achieve automatic generation of large-scale scenarios. It combines quantum-level waveform control and spacetime compression algorithms to break through the limitations of experimental scale and time. It uses intelligent algorithms to actively identify unconventional parameter changes, which significantly improves the comprehensiveness and efficiency of extreme working condition verification.

[0082] Through the above technical solutions, this application can discover potential failure modes that traditional experimental systems cannot cover, such as the sudden change in device insulation performance caused by the synergistic effect of high-frequency oscillation lightning and high temperature, providing more complete verification data support for the reliability design of new power equipment. At the same time, it reduces the cost of manual intervention through automated processes and realizes the intelligent upgrade of extreme working condition verification.

[0083] This application further proposes a verification process for climate change by constructing a digital experimental universe containing different climate zones, combining climate models to predict the future trend of increasing extreme weather, simulating lightning strike patterns under different climate conditions through parallel extrapolation and waveform control technology, statistically analyzing lightning strike failure rates, and optimizing grounding materials and layout schemes.

[0084] Among them, the digital experimental universe refers to a virtual experimental environment constructed through parametric modeling technology. Specifically, it can be implemented by integrating climate zone characteristic parameters, equipment aging parameters, and lightning parameters using a scene parametric modeling module to simulate the operating status of equipment under real climate conditions. The curiosity-driven mechanism of deep reinforcement learning refers to an algorithmic framework that guides the system to actively explore unconventional scenarios by constructing a reward function. Specifically, it can be implemented by combining a neural network model with an exploration strategy to explore the nonlinear correlation between climate conditions and lightning phenomena. The spatiotemporal folding algorithm refers to a numerical calculation method that mathematically models the physical and chemical changes during the equipment aging process. Specifically, it can be implemented by combining a time compression algorithm with a multiphysics coupling model to accelerate the simulation process of the impact of long-term climate change on equipment.

[0085] Specifically, the system integrates environmental parameters from climate zones such as tropical rainforests, polar regions, and deserts through a scene parameterization modeling module. Combined with climate models, it predicts future trends in extreme weather and constructs a digital experimental universe covering multi-dimensional climate conditions. During the simulation, it actively explores special lightning phenomena under different climate conditions using a deep reinforcement learning mechanism. For example, it simulates changes in lightning characteristics caused by humidity changes in coastal areas. The spatiotemporal folding algorithm accelerates the aging process of equipment under long-term climate change. The system analyzes the correlation between the performance degradation of insulation materials and the lightning failure rate through a multi-physics coupling model. The quantum-level waveform synthesis module and the adaptive phase-locked feedback module work together to generate lightning waveforms that conform to climate characteristics. Combined with a digital twin pre-simulation unit, it verifies the protective performance of the grounding system under different corrosion rates.

[0086] In some specific implementations, when simulating the impact of sea-level rise on the grounding system of coastal cities, the system can automatically adjust the soil resistivity parameters and metal corrosion rate parameters in the scenario gene, generate lightning strike waveforms under high salt spray environment through waveform control technology, and use the cross-temporal fault chain inference module to analyze the failure process of grounding materials under accelerated corrosion conditions.

[0087] Compared with existing technologies, traditional methods rely on historical data under a single climate condition for simulation, which cannot dynamically reflect the long-term impact of climate change on lightning strike patterns. This solution constructs a multi-dimensional climate digital experimental universe and combines an active exploration mechanism with a spatiotemporal compression algorithm to simultaneously simulate the coupling effect between climate evolution and equipment aging, thereby achieving full life cycle reliability assessment.

[0088] Through the above technical solution, this application effectively solves the problem that traditional experimental systems have difficulty predicting the impact of long-term climate change on the lightning protection performance of power equipment. By dynamically adjusting the environmental parameters and equipment status parameters in the digital experimental universe, it accurately simulates the corrosion rate changes of the grounding system under different climatic conditions, provides data support for optimizing the design of lightning protection devices, and avoids the risk of reduced protection performance due to climate change.

[0089] This application further proposes a verification process for the space environment by constructing an experimental scenario through a scenario parameterized modeling module that calls relevant waveform and parameter data of plasma lightning. The quantum-level waveform synthesis module, adaptive phase-locked feedback module, and physical generation feedback correction unit work together to generate a waveform containing particle bombardment through waveform control technology. This waveform is then applied to a digital twin pre-simulation unit for simulation verification to test the effectiveness of the protection design and solve the problem that traditional ground experiments cannot reproduce extreme space scenarios.

[0090] Among them, the scene parameterization modeling module refers to the digital modeling unit used to construct experimental scenarios. Specifically, it can be implemented using multi-source data fusion and parameter encoding technology. It generates high-fidelity experimental scenarios by integrating the waveform characteristics of plasma lightning with space environment parameters. The quantum-level waveform synthesis module refers to the device that generates high-precision waveforms based on quantum physics principles. Specifically, it can be implemented using a Josephson junction array voltage source combined with a quantum random number generator. It generates picosecond-level ultrashort pulses by suppressing waveform jitter. The adaptive phase-locked feedback module refers to the closed-loop control unit used to adjust waveform parameters in real time. Specifically, it can be implemented using deep learning neural networks and FPGA hardware acceleration technology. It maintains waveform stability through feature extraction and parameter optimization. The physical generation feedback correction unit refers to the actuator that connects the digital model and physical equipment. Specifically, it can be implemented using a quantum electric field sensor and a backpropagation algorithm. It corrects model parameters through real-time data comparison. The digital twin pre-simulation unit refers to the simulation platform based on multi-physics coupling. Specifically, it can be implemented using finite element analysis and Monte Carlo simulation methods. It verifies the effectiveness of the protection design through virtual-real mapping.

[0091] Specifically, the scene parameterization modeling module first calls the waveform parameters of plasma lightning and space environment data to construct an experimental scene that includes particle bombardment characteristics. The quantum-level waveform synthesis module generates a high-precision pulse waveform according to the scene parameters. At the same time, the adaptive phase-locked feedback module monitors the waveform status in real time and maintains the waveform stability by adjusting the capacitance and inductance parameters. The physical generation feedback correction unit converts the waveform instructions output by the digital model into physical signals, driving the actual generator to output a composite waveform containing particle bombardment characteristics. In the digital twin pre-simulation unit, this waveform is applied to the digital twin of the space equipment. The ability of the protective structure to withstand extreme environments is verified through multi-physics coupling simulation. When insulation breakdown or electromagnetic interference anomalies are detected, the system automatically corrects the model parameters and regenerates the optimized experimental waveform, forming a closed-loop verification process.

[0092] Compared with existing technologies, traditional ground experiments are limited by physical conditions and cannot simulate the particle bombardment effect in the space environment, resulting in blind spots in the verification of protection design. This solution combines quantum-level waveform synthesis and digital twin technology to accurately reproduce the electromagnetic characteristics of extreme space scenarios in a laboratory environment. At the same time, it uses an adaptive feedback mechanism to ensure the authenticity and stability of waveform generation, solving the problem of mismatch between experimental scenarios and real working conditions.

[0093] Through the above technical solution, this application achieves efficient verification of the protective structure of space equipment, and can accurately assess its reliability in extreme electromagnetic environments under laboratory conditions. It makes up for the technical deficiency of traditional experimental methods that cannot simulate particle bombardment effects, and provides an effective testing method for the lightning protection design of space equipment.

[0094] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A system for constructing a judgment model based on lightning impact experimental data using digital twins, characterized in that, include: The digital twin pre-simulation unit is used to simulate the impact of the target waveform on the equipment under different environmental conditions based on finite element analysis and multiphysics coupling model. It predicts the parameters required to generate the target waveform through Monte Carlo simulation method and transmits the parameters to the quantum-level waveform synthesis module and the physical generation feedback correction unit at the same time. The quantum-level waveform synthesis module is used to combine the waveform jitter suppression algorithm of the quantum random number generator to generate pulses using the received parameters, forming a waveform as the basis of the signal source; The physical generation feedback correction unit is used to drive the physical waveform generator to output waveforms and uses a quantum electric field sensor to collect waveform data in real time. The collected data is compared with the target waveform of the digital twin pre-simulation unit, and the digital model parameters are corrected through the back propagation algorithm to form closed-loop control. The adaptive phase-locked feedback module is used to acquire target waveform and actual output waveform data in real time. It uses a waveform tracking neural network based on deep learning to extract features and perform comparative analysis to determine whether there is amplitude attenuation or phase shift anomaly.

2. The lightning impact experimental data judgment model construction system based on digital twin as described in claim 1, characterized in that: The digital twin pre-simulation unit, together with the quantum-level waveform synthesis module, the physical generation feedback correction unit, and the adaptive phase-locked feedback module, constitutes a waveform control unit, which is used for waveform restoration processing.

3. The lightning impact experimental data judgment model construction system based on digital twin as described in claim 2, characterized in that: The waveform restoration process is as follows: S1: Predict the parameters required to generate the target waveform through the digital twin pre-simulation unit, and transmit these parameters simultaneously to the quantum-level waveform synthesis module and the physical generation feedback correction unit; S2: The quantum-level waveform synthesis module combines the waveform jitter suppression algorithm of the quantum random number generator to generate pulses using the received parameters, forming a waveform as the basis of the signal source; S3: The physical generation feedback correction unit drives the physical waveform generator to output the initial physical waveform based on the parameters provided by the digital twin pre-simulation unit; S4: The physical generation feedback correction unit uses a quantum electric field sensor to collect waveform data in real time, compares it with the results of the digital twin pre-simulation, and simultaneously synchronizes the real-time state of the waveform to the adaptive phase-locked feedback module. When the physical generation feedback correction unit finds that there is a deviation between the collected waveform data and the parameters generated by the digital twin pre-simulation unit, it corrects the digital model parameters through the backpropagation algorithm and feeds back the optimized parameters to the digital twin pre-simulation unit. S5: The digital twin pre-simulation unit re-evaluates the waveform generation strategy based on the new parameters, and then sends the adjusted instructions to the quantum-level waveform synthesis module and the adaptive phase-locked feedback module. S6: The quantum-level waveform synthesis module adjusts the voltage source output according to the new parameters, and the adaptive phase-locked feedback module automatically adjusts the capacitance and inductance parameters of the pulse generator through the FPGA.

4. The lightning impact experiment data judgment model construction system based on digital twin as described in claim 1, characterized in that: The system also includes a scene parametric modeling module for constructing a digital experimental universe, the specific process of which is as follows: A1: Obtain non-standard waveform data from the quantum-level waveform synthesis module and the adaptive phase-locked feedback module, and use it as the basis for waveform parameters in the scene gene fragment; A2: Use the parameters generated by the digital twin pre-simulation unit during the simulation process as the core basis for lightning parameters and equipment response parameters in the scene gene; A3: Based on real-time monitoring data from the adaptive phase-locked feedback module, the scene gene fragments are dynamically corrected.

5. The lightning impact experiment data judgment model construction system based on digital twin as described in claim 4, characterized in that: The system also includes an automatic exploration module for counterintuitive scenarios, which introduces a curiosity-driven mechanism based on deep reinforcement learning, constructs a reward function to guide the active exploration of unconventional scenarios in the digital experimental universe, and triggers exploration behavior when the system detects that the combination of environmental parameters deviates from the cognitive range.

6. The lightning impact experimental data judgment model construction system based on digital twin as described in claim 5, characterized in that: The system also includes a cross-temporal fault chain simulation module, which uses a spatiotemporal folding algorithm to mathematically model the physical and chemical changes during the equipment aging process. Through numerical calculation methods, it simulates the lightning impact cycle of the transformer, and can completely deduce the entire life cycle evolution process of its insulation material from crack initiation to breakdown, thereby locating the critical point of cumulative effect.

7. The lightning impact experimental data judgment model construction system based on digital twin as described in claim 6, characterized in that: The automatic exploration module for counterintuitive scenarios is also used for verification processing of extreme working conditions, climate change, and space environment.

8. The lightning impulse experimental data judgment model construction system based on digital twin as described in claim 7, characterized in that: The verification process for the extreme operating conditions is as follows: Based on the finite element analysis and multiphysics coupling model of the digital twin pre-simulation unit, a digital twin of the device under unconventional operating conditions is constructed. The spatiotemporal folding algorithm of the cross-spatiotemporal fault chain inference module is used to compress and simulate the device aging process; By leveraging the deep reinforcement learning mechanism of the counterintuitive scenario automatic exploration module, combined with a preset failure threshold, it proactively identifies deviations from conventional performance parameter changes, ultimately uncovering potential failure modes that traditional experiments cannot cover.

9. A system for constructing a lightning impact experimental data judgment model based on digital twins according to claim 7, characterized in that: The verification process for the climate change is as follows: A digital experimental universe containing different climate zones was constructed using the scene parametric modeling module; By introducing a curiosity-driven mechanism based on deep reinforcement learning, a reward function is constructed to guide active exploration of unconventional scenarios. Potential lightning phenomena and risks are explored under different climatic conditions. Using a spatiotemporal folding algorithm, the physical and chemical changes of equipment during the aging process under different climatic conditions are mathematically modeled. Through numerical calculation methods, the entire life cycle evolution process of equipment under the influence of long-term climate change is simulated, the critical point of cumulative effect is located, and the future weather is predicted by combining digital experimental universe. The lightning strike patterns under different climatic conditions are simulated, and the lightning failure rate is statistically analyzed.

10. A lightning impact experimental data judgment model construction system based on digital twin as described in claim 7, characterized in that: The verification process for the space environment is as follows: The experimental scenario is constructed by calling the relevant waveform and parameter data of plasma lightning through the scenario parameterization modeling module. The quantum-level waveform synthesis module, adaptive phase-locked feedback module and physical generation feedback correction unit work together to generate a waveform containing particle bombardment through waveform control technology. The waveform is then applied to the digital twin pre-simulation unit for simulation verification to test the effectiveness of the protection design.

Images