Power grid dispatching training method and system based on panoramic digital twinborn simulation

Abstract

The invention discloses a power grid dispatching training method and system based on panoramic digital twinborn simulation, and relates to the technical field of power grid dispatching training, and the method comprises the steps: integrating a power grid topological structure, equipment operation data and historical fault data, constructing a three-level dynamic simulation model of an equipment level, a station level and a region level, and carrying out the simulation of the three levels based on the obtained real-time operation data. After dimensionality reduction and anomaly detection processing are conducted on the data, parameters and states of a three-level dynamic simulation model are updated, simulation is driven through the updated model, power flow analysis, load prediction and stability evaluation are combined, disaster scenes are generated, cognitive process data are collected, quantitative analysis is conducted in combination with preset evaluation indexes, and personalized training feedback is generated. The power grid operation state is highly restored, the data accuracy is improved, the training content is enriched, the training effect is enhanced, the ability of dispatchers to deal with complex conditions is improved, the regional limitation is broken, and the training efficiency and the resource utilization rate are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid dispatching training, and particularly relates to a power grid dispatching training method and system based on panoramic digital twin simulation. BACKGROUND

[0002] In the field of power grid dispatching training, traditional training methods have many limitations. On the one hand, traditional training relies on theoretical explanation and simple simulation, which is difficult to truly restore the complex and variable actual operation scene of the power grid. Dispatchers are difficult to obtain a highly similar experience to actual work in training, resulting in a disconnection between training effectiveness and actual work needs. For example, in the face of complex situations such as extreme weather and new energy fluctuations, traditional training cannot fully understand the impact mechanism of the power grid and how to effectively respond.

[0003] On the other hand, with the continuous expansion of the power grid scale and the increasing complexity of the structure, the amount of power grid operation data is growing explosively. The accuracy, real-time performance and efficient processing of data have become key issues. Existing technologies are difficult to comprehensively and accurately collect and process these data, resulting in deviations in the data basis on which the training process is based, which further affects the accuracy and reliability of the digital twin model. For example, the data collected by some sensors has noise, which will cause errors in model construction if not effectively processed, and cannot truly reflect the actual operation state of the power grid.

[0004] In addition, traditional training is also relatively single in training evaluation, only focusing on the operation results of dispatchers and ignoring the cognitive process data in the operation process. It is difficult to deeply analyze the decision logic and thinking mode of dispatchers and to provide targeted improvement suggestions and optimize the training scheme, which is not conducive to the improvement of the comprehensive quality and business ability of dispatchers. SUMMARY

[0005] In view of the above existing problems, the present application provides a power grid dispatching training method and system based on panoramic digital twin simulation, to solve the problem that it is difficult to truly restore the complex and variable actual operation scene of the power grid in the prior art, and dispatchers are difficult to obtain a highly similar experience to actual work in training, resulting in a disconnection between training effectiveness and actual work needs.

[0006] To solve the above technical problems, a power grid dispatching training method based on panoramic digital twin simulation is proposed, which comprises,

[0007] The integrated power grid topology, equipment operation data and historical fault data are used to construct a three-level dynamic simulation model of the equipment level, station level and regional level; based on the obtained real-time operation data, the parameters and state of the three-level dynamic simulation model are updated after dimension reduction and abnormality detection processing of the data; the updated model is used to drive simulation, combined with power flow analysis, load prediction and stability evaluation, to generate a disaster scenario, collect cognitive process data, and generate personalized training feedback through quantitative analysis combined with preset evaluation indexes.

[0008] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the three-level dynamic simulation model of the equipment level, station level and regional level is constructed by collecting power grid current, voltage and power data through multi-source sensors, performing fusion and noise filtering processing on the data, and analyzing node connection relationship and clustering characteristics based on power grid topology, combined with device physical characteristics to construct an equipment level insulation aging and mechanical wear model, a station level three-dimensional real scene electromagnetic interference model and a regional level power grid flow-stability coupling model.

[0009] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the parameters and state of the three-level dynamic simulation model are updated by obtaining real-time data from the power grid monitoring system through an encryption transmission protocol, performing dimension reduction processing on multi-dimensional data, identifying data deviation using an abnormality analysis algorithm, optimizing model parameter update strategy based on current operating state and prediction data, and synchronizing actual power grid dynamic characteristics.

[0010] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the abnormality analysis algorithm includes calculating the depth deviation value of the data points in the path isolation tree, identifying abnormal data points that exceed the threshold range, and performing correction and rejection on the abnormal data to update the model parameters.

[0011] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the generation of the disaster scenario includes generating an extreme weather scenario based on historical disaster data, and simulating the fault evolution process of different severity by adjusting the disaster intensity, duration and influence range parameters, and dynamically simulating the line fault and voltage fluctuation caused by typhoon, blizzard and new energy fluctuation in combination with the real-time power grid data driven simulation environment.

[0012] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the generation of the disaster scenario further includes inputting typhoon path, wind speed and power grid characteristic data into a disaster prediction model, calculating the line fault probability, and simulating the force and electromagnetic interference effect of extreme weather on equipment based on fluid mechanics and electromagnetism theory.

[0013] An immersive training environment is constructed, and the dispatcher views real-time parameter changes by gesture operation of virtual equipment, and device-level fault phenomena and regional-level power grid cascading reactions are dynamically displayed based on three-dimensional real scene modeling;

[0014] The model prediction result is combined with the power grid model to simulate the influence of the typhoon on the power grid, and the probability of line fault is represented as:

[0015] P(fault)=σ(W2σ(W1X+b1)+b2)

[0016] Wherein, W1 and W2 are weight matrices, b1 and b2 are bias vectors, sigma is an activation function, and P(fault) is the probability of line fault.

[0017] As a preferred scheme of the power grid dispatching training method based on panoramic digital twin simulation, the generation of personalized training feedback includes recording the operation steps, decision-making time and state judgment data of the dispatcher in the disaster scene, constructing a multi-factor evaluation index system, quantifying the operation accuracy and decision-making efficiency through fuzzy synthesis operation, and generating skill short board improvement suggestions based on the quantization results and cognitive data analysis;

[0018] The construction of the multi-factor evaluation index system includes setting the weights of the fault handling speed and the load adjustment rationality index through the analytic hierarchy process, and synthesizing the weight vector and the fuzzy evaluation matrix to output the comprehensive ability score and the decision logic optimization path.

[0019] As a preferred scheme of the power grid dispatching training system based on panoramic digital twin simulation, it is characterized by comprising a data acquisition module, a digital twin model construction module, a model updating module, a disaster scene simulation module, a training evaluation module, an interactive training module and a cloud collaboration module.

[0020] The data acquisition module is used to acquire real-time operation data of the power grid through current sensors, voltage sensors and power sensors, and to process noise data by using Kalman filtering algorithm to provide high-precision data basis for the digital twin model.

[0021] The digital twin model construction module is used to construct a three-level dynamic simulation model based on the power grid topology structure and the device physical characteristics.

[0022] The model updating module is used to acquire real-time data from the power grid monitoring system through AES encryption algorithm, to process data by using principal component analysis dimension reduction, and to calculate data point path depth deviation value by using isolated forest algorithm to detect abnormal data, and to update model parameters by model prediction control strategy.

[0023] The disaster scenario simulation module is configured to generate typhoon and snowstorm scenarios based on historical disaster data by using a Monte Carlo simulation method, calculate line fault probability by using a convolutional neural network, simulate wind turbine blade stress by using fluid mechanics, calculate power transmission line interference by using electromagnetism theory, and adjust disaster intensity to simulate different severity faults.

[0024] The training evaluation module is configured to record dispatcher operation steps and decision-making time, establish a judgment matrix by using an analytic hierarchy process, calculate index weight, construct an evaluation matrix by using a fuzzy comprehensive evaluation method, and perform synthetic operation to generate a quantitative evaluation result and improvement suggestions.

[0025] The interactive training module is configured to integrate VR and AR technologies to construct a three-dimensional real scene power grid environment, support dispatchers to view real-time parameter changes by gesture operating virtual equipment, and dynamically display equipment-level fault phenomena and regional-level cascading reactions.

[0026] The cloud-based collaboration module is configured to deploy distributed storage and calculation modules based on a cloud computing platform, and realize data synchronization, multi-user remote collaborative training and joint examination of cross-regional dispatching centers by using WebRTC real-time communication technology.

[0027] A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method for power grid dispatching training based on panoramic digital twin simulation when executing the computer program.

[0028] A computer readable storage medium stores a computer program, and the computer program implements the steps of the method for power grid dispatching training based on panoramic digital twin simulation when executed by a processor.

[0029] The present application has the following beneficial effects: the present application integrates power grid topology, equipment operation data and historical fault data, constructs a three-level dynamic simulation model including equipment level, station level and regional level, realizes full-dimensional accurate restoration from micro device physical characteristics to macro power grid operation state, and solves the problem of low restoration degree of traditional training scene; through multi-source sensor data fusion and Kalman filter noise processing, combined with topology structure node degree and clustering coefficient analysis, the high precision and topology logic consistency of model input data are ensured. The model parameters are dynamically updated by using encrypted real-time data acquisition, PCA dimension reduction and isolation forest anomaly detection technology, so that the response delay of the digital twin model is reduced to seconds, and the dynamic changes of the actual power grid are synchronized; based on Monte Carlo simulation and CNN fault probability prediction, a gradient disaster scene with adjustable parameters is generated, combined with fluid mechanics and electromagnetic theory to simulate the physical impact of extreme weather on equipment, realizing the intelligent training of "randomness-probability-physicality" three realness guarantee; through VR gesture interaction three-dimensional display of fault chain reaction, AHP dynamic weight and fuzzy comprehensive evaluation are used to quantitatively analyze the cognitive process data of dispatchers, forming a "operation-evaluation-optimization" self-evolution training closed loop, and constructing a full-chain power grid dispatching meta-universe training system covering data acquisition, simulation modeling, dynamic updating, scene generation and cognitive evaluation, which significantly improves the authenticity, pertinence and systematicness of the training. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0031] Figure 1 The overall flowchart of a power grid dispatching training method based on panoramic digital twin simulation provided by an embodiment of the present application.

[0032] Figure 2 The real-time updating panoramic digital twin model flowchart of a power grid dispatching training method based on panoramic digital twin simulation provided by an embodiment of the present application.

[0033] Figure 3 The system scheme flowchart of a power grid dispatching training system based on panoramic digital twin simulation provided by an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present application.

[0035] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. The present application, however, can be practiced in a variety of ways beyond the specific details set forth herein without departing from the scope of the present application, and it is understood that the present application is not limited to the particular details described. Accordingly, the particular shape of the described devices can not be drawn to scale and the drawings are shown for purposes of explanation only.

[0036] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor does it mean that the embodiment is mutually exclusive with other embodiments.

[0037] The present application is described in detail in conjunction with the schematic drawings. In the detailed description of the embodiments of the present application, the cross-sectional view of the device structure is locally enlarged without the general proportion for the convenience of description, and the schematic drawings are only examples which should not limit the scope of protection of the present application. In addition, three-dimensional spatial dimensions including length, width and depth should be included in actual manufacturing.

[0038] Meanwhile, in the description of the present application, it should be noted that the terms "upper, lower, inner and outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first, second or third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0039] Unless otherwise specifically defined and limited, the terms "mounting, connecting, connection" in the present application should be understood broadly, for example: it can be fixed connection, detachable connection or integral connection; it can also be mechanical connection, electrical connection or direct connection, it can also be indirectly connected through intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0040] Example 1, refer to Figure 1 and Figure 2For the first embodiment of the present application, the embodiment provides a power grid dispatching training method based on panoramic digital twin simulation, comprising:

[0041] S1: integrate power grid topology, equipment operation data and historical fault data to construct a three-level dynamic simulation model at the device level, station level and regional level.

[0042] Further, the three-level dynamic simulation model at the device level, station level and regional level comprises collecting power grid current, voltage and power data through multi-source sensors, fusing and filtering the data, analyzing node connection relationship and clustering characteristics based on power grid topology, and constructing device-level insulation aging and mechanical wear models, station-level three-dimensional real scene electromagnetic interference models and regional-level power flow-stability coupling models in combination with equipment physical properties.

[0043] The three-level dynamic simulation model at the device level, station level and regional level further comprises data collection, data integration, mathematical model establishment and simulation algorithm, and realization of digital twin model construction; data collection comprises collecting power grid topology information, equipment operation data and historical fault data, using network crawler technology to obtain data from power grid database data source, and preliminarily cleaning and arranging the data;

[0044] Data integration comprises integrating real-time operation data including current, voltage and power, using a data fusion algorithm, i.e., a weighted average fusion method, to fuse the same type of data from different sensors, and the formula is:

[0045]

[0046] Wherein, I i is the current data collected by the sensor, I is the fused current value, n is the total number of sensors, i is the variable index, w i is the weight of the sensor data, which needs to satisfy

[0047] The construction of the digital twin model comprises using a three-level dynamic simulation model, at the device level, modeling the transformer and circuit breaker equipment using finite element analysis method to simulate the electromagnetic and thermal forces inside the equipment; at the station level, modeling the three-dimensional real scene and calculating the electromagnetic interference between equipment based on electromagnetic field theory; at the regional level, using power system analysis theory to perform power flow-stability coupling simulation to support N-2 fault under cascading reaction deduction.

[0048] Further, the noise filtering process comprises collecting power grid data using current, voltage and power sensors, in a complex power grid environment, the sensor data has noise, and the collected data is processed using Kalman filtering algorithm, and the formula is:

[0049]

[0050] P(k|k-1)=AP(k-1|k-1)A T +Q

[0051] K(k)=P(k|k-1)H T [HP(k|k-1)H T +R] -1

[0052]

[0053] P(k|k)=[F-K(k)]P(k|k-1)

[0054] wherein, is an estimated value of the voltage data based on information at time k-1 at time k, A is a state transition matrix, representing the change relationship of the system state from the last time to the current time, is the best estimated value of the voltage data at time k-1, B is a control matrix, representing the influence of the control input on the system state, u(k-1) is a control quantity, that is, the control input applied at time k-1, P(k|k-1) is a prediction error covariance matrix, representing the prediction uncertainty based on the information at time k-1 at time k, Q is a process noise covariance matrix, K(k) is a Kalman gain matrix, used to adjust the prediction value to consider new observation data, H is an observation matrix, representing how to get the observation data from the system state, T represents the transposition operation, R is an observation noise covariance matrix, representing the uncertainty of the observation data, is an updated estimated value of the voltage data at time k, F is an identity matrix, used to maintain the consistency of matrix operations, P(k|k) is an updated error covariance matrix, representing the estimation of uncertainty after considering new observation data.

[0055] S2: based on the obtained real-time running data, after dimensionality reduction and abnormality detection processing of the data, updating the parameters and states of the three-level dynamic simulation model.

[0056] Further, the updating of the parameters and states of the three-level dynamic simulation model includes obtaining real-time data from the power grid monitoring system through an encrypted transmission protocol, performing dimensionality reduction processing on the multi-dimensional data, identifying data deviation using an abnormality analysis algorithm, optimizing the model parameter update strategy based on the current running state and prediction data, and synchronizing the actual power grid dynamic characteristics.

[0057] Further, the updating of the parameters and states of the three-level dynamic simulation model further comprises obtaining real-time operation data, processing and analyzing the obtained data, and updating the parameters and states in the digital twin model; the obtaining of the real-time operation data comprises establishing a data transmission security mechanism, and using an encryption algorithm (AES encryption algorithm) to encrypt the transmission data to ensure data security.

[0058] The processing and analysis of the obtained data comprises using a principal component analysis (PCA) algorithm to reduce the dimensionality of the multi-dimensional data, reducing data redundancy, and improving data processing efficiency, projecting the original data to a low-dimensional space by selecting a set of orthogonal basis vectors through the PCA algorithm, and obtaining the reduced data.

[0059] The updating of the parameters and states in the digital twin model comprises using a model predictive control (MPC) algorithm to optimize the model parameter update strategy based on the current power grid operation data and the predicted future data, so that the model better tracks the actual power grid operation state.

[0060] The anomaly analysis algorithm comprises calculating the depth deviation value of the data points in the path isolation tree, identifying abnormal data points that exceed the threshold range, and performing correction and elimination on the abnormal data to update the model parameters.

[0061] The updating of the model parameters comprises using data mining techniques to process and analyze the data, using an isolation forest algorithm to calculate the outlier score of each data point, and judging whether the data is abnormal according to the score, and updating the related parameters and states in the digital twin model after correcting or eliminating the abnormal data to ensure that the model is synchronized with the actual power grid operation state, and the formula is represented as:

[0062]

[0063] wherein h(x) is the path length of the data point x to the root node in the isolation forest, S(x) is the outlier score of the data point x, E(h(x)) is the expected value of the path length, and c(m) is a constant related to the number of samples m.

[0064] In the embodiments of the present application, the updating of the model parameters comprises obtaining real-time current, voltage, and power data from the power grid monitoring system through the AES encryption protocol; using principal component analysis (PCA) to reduce the dimensionality of the multi-dimensional data and retain the principal component features; using an isolation forest algorithm to calculate the path depth deviation value of the data points and identify abnormal data; and using model predictive control (MPC) to optimize the model parameter update strategy to synchronize the digital twin model with the actual power grid state.

[0065] In an alternative embodiment, updating the model parameters includes protecting real-time data transmission by differential privacy encryption technology (such as Laplace noise injection), using an automatic encoder (Autoencoder) for nonlinear dimensionality reduction of high-dimensional data, reconstructing key features, using robust regression (Huber loss function) to detect data outliers, reducing the influence of outliers on the model, and dynamically adjusting the model parameters based on a sliding window incremental learning algorithm (online gradient descent) to adapt to power grid fluctuations.

[0066] In another alternative embodiment, updating the model parameters includes using homomorphic encryption (Paillier algorithm) to implement ciphertext data calculation to ensure data security, using t-SNE (t-distributed stochastic neighbor embedding) for nonlinear dimensionality reduction to preserve local data structure, constructing a normal data distribution boundary through a One-Class SVM algorithm to identify abnormal data points, and applying an adaptive Kalman filter (AKF) to dynamically update the model parameters to optimize state estimation accuracy.

[0067] S3: Using the updated model to drive simulation, combining power flow analysis, load forecasting, and stability evaluation to generate disaster scenarios, collect cognitive process data, and conduct quantitative analysis based on preset evaluation indicators to generate personalized training feedback.

[0068] Further, the generation of disaster scenarios includes generating extreme weather scenarios based on historical disaster data, and simulating fault evolution processes of different severity by adjusting disaster intensity, duration, and impact range parameters, and dynamically simulating line faults and voltage fluctuations caused by typhoons, snowstorms, and new energy fluctuations in a real-time power grid data-driven simulation environment.

[0069] The generation of disaster scenarios also includes inputting typhoon path, wind speed, and power grid characteristic data into a disaster prediction model to calculate line fault probability, and simulating the force of extreme weather on equipment and electromagnetic interference effects based on fluid mechanics and electromagnetism theory.

[0070] In the embodiments of the present application, the generation of disaster scenarios includes collecting historical typhoon, snowstorm disaster data and power grid fault records, using Monte Carlo simulation to randomly generate disaster event sequences, predicting line fault probability through a CNN model, and simulating equipment physical response based on fluid mechanics and electromagnetism theory.

[0071] In an alternative embodiment, the generation of disaster scenarios includes synthesizing extreme weather data using GAN (Generative Adversarial Network) to expand training samples, and calculating fault condition probability using Bayesian network, simulating the impact of wind turbine mechanical vibration on the power grid based on multi-body dynamics simulation (ADAMS software), and coupling electromagnetic transient simulation (EMTP) to analyze fault current propagation path.

[0072] In another alternative embodiment, the generating disaster scenarios includes predicting the typhoon path and the temporal variation of intensity using a Transformer model, simulating the autonomous response behavior of power grid equipment in disasters based on Agent modeling, simulating the mechanical load of snow on power transmission lines using the discrete element method (DEM), and calculating the stress distribution of the tower structure by combining finite element analysis (FEA).

[0073] An immersive training environment is constructed, and the operator views real-time parameter changes by gesture operation of virtual equipment, and the equipment-level fault phenomenon and regional-level power grid cascading reaction are dynamically displayed based on three-dimensional real scene modeling.

[0074] The model prediction results are combined with the power grid model to simulate the impact of typhoons on the power grid, and the probability of line failure is represented as:

[0075] P(fault)=σ(W2σ(W1X+b1)+b2)

[0076] Where W1 and W2 are weight matrices, b1 and b2 are bias vectors, σ is the activation function, and P(fault) is the line failure probability.

[0077] Further, the generation of personalized training feedback includes recording the operation steps, decision-making time and state judgment data of the operator in the disaster scenario, constructing a multi-factor evaluation index system, quantifying the operation accuracy and decision-making efficiency through fuzzy synthesis operation, and generating skill short board improvement suggestions based on the quantization results and cognitive data analysis;

[0078] The construction of the multi-factor evaluation index system includes setting the weights of the fault handling speed and load adjustment rationality indicators through the analytic hierarchy process, and synthesizing the weight vector and the fuzzy evaluation matrix to output the comprehensive ability score and the decision logic optimization path, which is represented by the formula:

[0079] JO=λ max O

[0080] Where J is the judgment matrix, O is the characteristic vector, and λ max is the largest eigenvalue of the judgment matrix.

[0081] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

[0082] Example 2, refer to Figure 3For the second embodiment of the present application, the embodiment provides a power grid dispatching training system based on panoramic digital twin simulation, comprising a data acquisition module, a digital twin model construction module, a model updating module, a disaster scenario simulation module, a training evaluation module, an interactive training module, and a cloud collaboration module.

[0083] The data acquisition module is configured to acquire real-time operation data of the power grid through current sensors, voltage sensors, and power sensors, and process the noisy data by using a Kalman filtering algorithm to provide a high-precision data basis for the digital twin model.

[0084] The digital twin model construction module is configured to construct a three-level dynamic simulation model based on the power grid topology and the physical characteristics of the equipment.

[0085] The model updating module is configured to obtain real-time data from the power grid monitoring system by using an AES encryption algorithm, process the data by using principal component analysis dimension reduction, calculate the data point path depth deviation value by using an isolation forest algorithm, detect abnormal data, and update the model parameters by using a model predictive control strategy.

[0086] The disaster scenario simulation module is configured to generate typhoon and snowstorm scenarios by using a Monte Carlo simulation method based on historical disaster data, calculate the line fault probability by using a convolutional neural network, simulate the stress on the fan blades by using fluid mechanics, calculate the power transmission line interference by using electromagnetism theory, and adjust the disaster intensity to simulate different severity faults.

[0087] The training evaluation module is configured to record the operation steps and decision-making time of the dispatchers, establish a judgment matrix by using an analytic hierarchy process, calculate the index weight, construct an evaluation matrix by using a fuzzy comprehensive evaluation method, and perform a synthesis operation to generate a quantitative evaluation result and improvement suggestions.

[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and they should be covered in the scope of the claims of the present application.

[0089] In embodiment 3, the third embodiment of the present application is different from the first two embodiments in that:

[0090] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0091] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered a list of executable instructions for implementing logic functions, and can be specifically embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or in conjunction with these instructions execution systems, apparatuses, or devices. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport programs for use by an instruction execution system, apparatus, or device, or in conjunction with these instruction execution systems, apparatuses, or devices.

[0092] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by editing, interpreting, or otherwise processing, if necessary, in other suitable ways to be electronically obtained, and then stored in the computer memory.

[0093] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the embodiments described above, various steps or methods can be implemented, for example, by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following technology, known in the art, or combinations thereof, can be used: discrete logic circuitry having logic gates for implementing logic functions upon an application of data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

Claims

1. A power grid dispatch training method based on panoramic digital twin simulation, characterized in that: The application relates to a method for generating individualized training feedback for power grid dispatchers. The method comprises the following steps: collecting power grid current, voltage and power data through multi-source sensors, fusing and filtering the data, analyzing node connection relationships and clustering characteristics based on power grid topology, and constructing a device-level insulation aging and mechanical wear model, a station-level three-dimensional real scene electromagnetic interference model and a regional power grid power flow-stability coupling model in combination with device physical characteristics. The method further comprises the following steps: obtaining real-time data from a power grid monitoring system through an encryption transmission protocol, performing dimensionality reduction processing on multi-dimensional data, identifying data deviations through an abnormality analysis algorithm, optimizing model parameter update strategies based on current operating states and prediction data, and synchronizing actual power grid dynamic characteristics. The method further comprises the following steps: calculating the depth deviation value of a data point in a path isolation tree, identifying abnormal data points that exceed a threshold range, performing correction and elimination on the abnormal data, and updating model parameters.

2. The power grid dispatch training method based on panoramic digital twin simulation according to claim 1, characterized in that: The method further comprises the following steps: generating an extreme weather scenario based on historical disaster data, simulating fault evolution processes of different severity by adjusting disaster intensity, duration and influence range parameters, and dynamically simulating line faults and voltage fluctuations caused by typhoons, snowstorms and new energy fluctuations in a simulation environment driven by real-time power grid data.

3. The power grid dispatch training method based on panoramic digital twin simulation according to claim 2, characterized in that: The method further comprises the following steps: inputting typhoon path, wind speed and power grid characteristic data into a disaster prediction model, calculating line fault probability, and simulating the force and electromagnetic interference effect of extreme weather on equipment based on fluid mechanics and electromagnetism theories.

4. The power grid dispatch training method based on panoramic digital twin simulation according to claim 3, characterized in that: The method further comprises the following steps: constructing an immersive training environment, enabling dispatchers to view real-time parameter changes by operating virtual equipment through gestures, and dynamically displaying device-level fault phenomena and regional power grid cascading reactions based on three-dimensional real scene modeling.

5. The power grid dispatch training method based on panoramic digital twin simulation according to claim 4, characterized in that: The method further comprises the following steps: combining model prediction results with a power grid model, simulating the influence of a typhoon on a power grid, and expressing the probability of line faults as follows: P(fault) = sigma(W2sigma(W1X+b1)+b2), wherein W1 and W2 are weight matrices, b1 and b2 are bias vectors, sigma is an activation function, and P(fault) is the probability of line faults.

6. The power grid dispatch training method based on panoramic digital twin simulation according to claim 5, characterized in that: The method further comprises the following steps: recording the operation steps, decision-making time and state judgment data of dispatchers in disaster scenarios, constructing a multi-factor evaluation index system, quantifying operation accuracy and decision-making efficiency through fuzzy synthesis operation, and generating skill short board improvement suggestions based on quantification results and cognitive data analysis. ​ ​ ​ ​ 7. The power grid dispatch training method based on panoramic digital twin simulation according to claim 6, characterized in that: ​ The multi-factor evaluation index system is constructed, including setting weights of fault handling speed and load adjustment rationality indexes by an analytic hierarchy process, and synthesizing the weight vector and a fuzzy evaluation matrix to output a comprehensive capability score and a decision logic optimization path.

8. A system for power grid dispatch training using a panoramic digital twin simulation-based method according to any one of claims 1-7, characterized in that: The system comprises a data acquisition module, a digital twin model construction module, a model updating module, a disaster scenario simulation module, a training evaluation module, an interactive training module, and a cloud collaboration module. The data acquisition module is configured to acquire real-time operation data of the power grid through current sensors, voltage sensors, and power sensors, and process the noisy data by using a Kalman filtering algorithm to provide a high-precision data basis for the digital twin model. The digital twin model construction module is configured to construct a three-level dynamic simulation model based on a power grid topology structure and device physical characteristics. The model updating module is configured to acquire real-time data from a power grid monitoring system by using an AES encryption algorithm, process the data by using principal component analysis dimension reduction, calculate data point path depth bias values by using an isolation forest algorithm, detect abnormal data, and update model parameters by using a model predictive control strategy. The disaster scenario simulation module is configured to generate typhoon and snowstorm scenarios by using a Monte Carlo simulation method based on historical disaster data, calculate line fault probabilities by using a convolutional neural network, simulate wind turbine blade forces by using fluid mechanics, calculate power transmission line interference by using electromagnetism theory, and adjust disaster intensity to simulate different severity faults. The training evaluation module is configured to record dispatcher operation steps and decision times, establish a judgment matrix by using an analytic hierarchy process, calculate index weights, construct an evaluation matrix by using a fuzzy comprehensive evaluation method, and perform a synthesis operation to generate a quantitative evaluation result and improvement suggestions. The interactive training module is configured to integrate VR and AR technologies to construct a three-dimensional real scene power grid environment, support dispatchers to view real-time parameter changes by operating virtual devices through gestures, and dynamically display device-level fault phenomena and regional-level cascading reactions. The cloud collaboration module is configured to deploy distributed storage and calculation modules based on a cloud computing platform, and realize data synchronization, multi-user remote collaborative training, and joint assessment of cross-regional dispatching centers by using WebRTC real-time communication technology. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The computer program is executed by the processor to realize the steps of the power grid dispatching training method based on the panoramic digital twin simulation according to any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the power grid dispatching training method based on the panoramic digital twin simulation according to any one of claims 1 to 7.

Images