Simulation system for secondary equipment of power grid

Abstract

The invention provides a power grid secondary equipment analog simulation system, and relates to the technical field of power secondary equipment, and the system comprises a data collection module which is used for obtaining the power data of the power grid secondary equipment; the data processing module is used for performing data preprocessing on the power data to obtain sample training data; the simulation model construction module is used for capturing the complex dynamic characteristics of the power grid secondary equipment through a deep neural network by utilizing the sample training data and the physical characteristics of the power grid secondary equipment, and constructing a multi-level simulation model; the simulation execution module is used for automatically optimizing and adjusting the corresponding to-be-simulated key parameters according to the simulation basic parameters set by the user on the basis of the multi-level simulation model to obtain simulation result data; and the simulation result recording module is used for monitoring changes of the key parameters in real time in the simulation process and recording and displaying simulation result data. According to the scheme, the accuracy and reliability of analogue simulation of the power grid secondary equipment can be improved.

Description

Technical Field

[0001] The invention relates to the technical field of electric power secondary equipment, and in particular to a power grid secondary equipment simulation system. Background Art

[0002] Power secondary equipment is an auxiliary device that monitors, measures, controls, regulates and protects the primary equipment of the power system, and is not directly connected to the main circuit of the electric energy. At the same time, the secondary equipment of the power grid is also a key equipment used in the power system to realize the functions of protection, control, measurement and monitoring. Its reliability and accuracy directly affect the safety and stability of the power system.

[0003] With the development of smart grid and energy Internet, the power grid industrial control system has turned from relatively closed to open, and the interaction between the power grid industrial control system and external systems has been increasing. Therefore, it is necessary to combine the security protection requirements of the power grid industrial control system, establish a security laboratory, and carry out research on security threat monitoring technology, vulnerability mining, penetration security, intentional attack blocking technology, access control technology and infection system isolation control technology. All of these are based on simulation, including the simulation of power grid secondary equipment. Through the simulation of substation secondary equipment, the cost of security testing can be greatly reduced, and the parameters of power grid secondary equipment can be presented to the staff more conveniently and clearly.

[0004] However, with the increasing complexity of power grid systems, traditional simulation methods often have poor accuracy and reliability in simulation results when dealing with high-dimensional and high-complexity power grid secondary equipment, making it difficult to evaluate the actual operating conditions of power grid secondary equipment. Summary of the invention

[0005] In view of this, in order to address the above shortcomings, it is necessary to propose a power grid secondary equipment simulation system to improve the accuracy and reliability of the simulation of power grid secondary equipment.

[0006] The present invention provides a power grid secondary equipment simulation system, including: a data acquisition module, a data processing module, a simulation model building module, a simulation execution module and a simulation result recording module;

[0007] The data acquisition module is used to obtain power data of the secondary equipment of the power grid; wherein the power data at least includes current data and voltage data;

[0008] The data processing module is used to perform data preprocessing on the power data to obtain sample training data;

[0009] The simulation model building module is used to use the sample training data and the physical characteristics of the power grid secondary equipment to capture the complex dynamic characteristics of the power grid secondary equipment through a deep neural network to build a multi-level simulation model;

[0010] The simulation execution module is used to automatically optimize and adjust the corresponding key parameters to be simulated based on the multi-level simulation model according to the simulation basic parameters set by the user to obtain simulation result data;

[0011] The simulation result recording module is used to monitor the changes of key parameters in real time during the simulation process, and to record and display the simulation result data.

[0012] Preferably, the data processing module includes a normalization processing unit, which performs normalization processing on the power data using the following calculation formula:

[0013]

[0014] Among them, X is the power data, X max and X min are the maximum and minimum values ​​in the power data, respectively, norm The data are normalized.

[0015] Preferably, the simulation model construction module includes a complex mathematical model construction unit, a multi-level simulation model construction unit and an iterative optimization unit;

[0016] The complex mathematical model construction unit is used to construct a complex network model of the following calculation formula through a deep neural network and a recursive neural network according to the sample training data and the physical characteristics of the power grid secondary equipment;

[0017] y t =σ(W h h t-1 +W x x t +b)

[0018] Among them, y t Represents the output at time t, x t Characterize the input data, h t-1 Represents the hidden state of the previous moment, W h and W x is the weight matrix, b is the bias parameter, and σ(·) is the activation function;

[0019] The multi-level simulation model construction unit is used to further construct a multi-level model based on a single deep neural network through the following calculation formula to capture the complex dynamic characteristics of the behavior of the secondary equipment of the power grid, and obtain a primary multi-level simulation model:

[0020]

[0021] Among them, h t (1) and h t (2) are the hidden states of the first and second layers respectively, W h (1) , W h (2) , W x (1) and is the weight matrix, b (1) , b (2) and b y is the bias parameter;

[0022] The iterative optimization unit is used to iteratively optimize the primary multi-level simulation model based on the loss function to obtain a multi-level simulation model.

[0023] Preferably, the iterative optimization unit is configured to perform the following operations when iteratively optimizing the primary multi-level simulation model based on the loss function:

[0024] S1: Initialize model parameters;

[0025] S2: In each iteration, the loss function is calculated based on the current model parameters;

[0026] S3: Calculate the gradient of the loss function and update the parameters according to the gradient descent method;

[0027] S4: Determine whether the updated parameters meet the convergence conditions; if so, stop the iteration and output the multi-level simulation model; otherwise, repeat steps S2-S4.

[0028] Preferably, the simulation system further includes a simulation platform building module, which includes a hardware platform building unit and a software platform building unit;

[0029] The hardware platform building unit is used to build a hardware platform using high-performance computing equipment or dedicated simulation hardware;

[0030] The software platform building unit is used to build a software platform that supports the integration and efficient calculation of complex models.

[0031] Preferably, the simulation execution module includes a parameter optimization adjustment unit, which is used to optimize and adjust the key parameters to be simulated through the following calculation formula:

[0032]

[0033] Among them, vi (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i at time t, p i is the individual optimal position of particle i, g is the global optimal position, ω is the inertia weight, c1 and c2 are acceleration constants, and r1 and r2 are random numbers.

[0034] Preferably, the simulation result recording module includes a simulation parameter monitoring unit and a simulation result display unit;

[0035] The simulation parameter monitoring unit is used to monitor the changes of key parameters in real time during the simulation process and record the simulation result data in detail; wherein the simulation result data includes the response time, load condition and fault status of the secondary equipment of the power grid;

[0036] The simulation result display unit is used to display the simulation result data through at least one of a chart, a report and a visualization tool.

[0037] Preferably, the simulation result recording module further includes an abnormal alarm unit;

[0038] The abnormal alarm unit is used to issue an alarm prompt when the simulation result data monitored by the simulation parameter monitoring unit is abnormal, so as to remind the staff to pay attention to the abnormality of the secondary equipment of the power grid.

[0039] Preferably, the simulation system further comprises: a simulation performance evaluation module;

[0040] The simulation performance evaluation module is used to compare and analyze the actual test results of the power grid secondary equipment and the simulation result data through the following calculation formula to verify the reliability of the multi-level simulation model:

[0041]

[0042] Among them, R i is the model error about the ith simulation parameter, Y j and are the actual values ​​of the power grid secondary equipment and the model simulation predicted values, respectively, and M is the type of simulation parameter.

[0043] Preferably, the simulation system further comprises: a model updating module;

[0044] The model updating module is used to determine the error R of the ith simulation parameter. i Is it greater than a preset threshold? If less than, the multi-level simulation model is updated with new power data until the error R of any simulation parameter is less than i are all smaller than the corresponding preset thresholds.

[0045] It can be seen from the above technical scheme that the power grid secondary equipment simulation system provided by the present invention includes a data acquisition module, a data processing module, a simulation model construction module, a simulation execution module and a simulation result recording module; first, the data acquisition module can obtain the power data of the power grid secondary equipment, and then the data processing module can perform data preprocessing on the power data to obtain sample training data; further, the simulation model construction module uses the sample training data and the physical characteristics of the power grid secondary equipment to capture the complex dynamic characteristics of the power grid secondary equipment based on the deep neural network, and constructs a multi-level simulation model; then the simulation execution module can perform simulation optimization and adjustment based on the basic parameters set by the user based on the multi-level simulation model to obtain the simulation result data, and the obtained simulation results are monitored, recorded and displayed by the simulation result recording module. It can be seen that this scheme can accurately describe the high-dimensional interaction and complex behavior characteristics of the power grid secondary equipment through the deep neural network, significantly improve the accuracy and reliability of the simulation model, and make the simulation results closer to the actual operating conditions. Moreover, during the simulation process, the simulation result recording module can monitor the changes of key erasers in real time and record the simulation result data in detail, which provides rich data support for subsequent simulation result analysis and model optimization, and helps to understand the behavior and performance characteristics of the power grid secondary equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of a power grid secondary equipment simulation system provided by an embodiment of the present invention.

[0047] Figure 2 A schematic diagram of a simulation model construction module provided by an embodiment of the present invention.

[0048] Figure 3 A schematic diagram of a simulation result recording module provided in an embodiment of the present invention.

[0049] Figure 4 A schematic diagram of a simulation platform building module provided in an embodiment of the present invention.

[0050] In the figure: data acquisition module 10, data processing module 20, simulation model construction module 30, complex mathematical model construction unit 31, multi-level simulation model construction unit 32, iterative optimization unit 33, simulation execution module 40, simulation result recording module 50, simulation parameter monitoring unit 51, simulation result display unit 52, abnormal alarm unit 53, simulation platform construction module 60, hardware platform construction unit 61, software platform construction unit 62, simulation performance evaluation module 70, model update module 80. DETAILED DESCRIPTION

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] See also Figure 1-4 The present invention provides a power grid secondary equipment simulation system, which may include: a data acquisition module 10, a data processing module 20, a simulation model building module 30, a simulation execution module 40 and a simulation result recording module 50;

[0053] The data acquisition module 10 is used to obtain power data of the secondary equipment of the power grid; wherein the power data at least includes current data and voltage data;

[0054] The data processing module 20 is used to perform data preprocessing on the power data to obtain sample training data;

[0055] The simulation model building module 30 is used to capture the complex dynamic characteristics of the power grid secondary equipment through a deep neural network using sample training data and the physical characteristics of the power grid secondary equipment to build a multi-level simulation model;

[0056] The simulation execution module 40 is used to automatically optimize and adjust the corresponding key parameters to be simulated based on the multi-level simulation model according to the simulation basic parameters set by the user to obtain simulation result data;

[0057] The simulation result recording module 50 is used to monitor the changes of key parameters in real time during the simulation process, and to record and display the simulation result data.

[0058] In this embodiment, the high-dimensional interactions and complex behavior characteristics of the power grid secondary equipment can be accurately described through the deep neural network, which significantly improves the accuracy and reliability of the simulation model, making the simulation results closer to the actual operating conditions. Moreover, during the simulation process, the simulation result recording module 50 can monitor the changes in key erase data in real time and record the simulation result data in detail, which provides rich data support for subsequent simulation result analysis and model optimization, and helps to understand the behavior and performance characteristics of the power grid secondary equipment. In addition, the data acquisition module 10 can obtain a variety of key operating data from the power grid in real time and accurately through the advanced data acquisition system, which provides a solid foundation for subsequent simulation analysis. Through data preprocessing, the accuracy and availability of the data are ensured, and the simulation errors caused by data quality problems are reduced.

[0059] The data acquisition module 10 is used to obtain power data of the secondary equipment of the power grid. For example, detailed operation data can be directly obtained from the actual equipment of the power grid based on the advanced data acquisition system. Specifically, the data acquisition module 10 may include a voltage sensor, a current sensor, and a memory, etc., to measure the battery voltage, current, power, frequency and other data of the secondary equipment of the power grid; for example, the collected voltage data is represented as V = {V1, V2, ..., V n}, where V n represents the nth voltage data, where n is a positive integer; the collected current data can be expressed as I = {I1, I2, ..., I m}, where I m Represents the mth current data, where m is a positive integer.

[0060] The data processing module 20 is used to pre-process the power data to obtain sample training data. Specifically, the data processing module 20 can remove noise, smooth data, normalize the power data, etc., to provide accurate data basis for subsequent modeling and simulation. For example, the data processing module 20 may include a normalization processing unit, which may use the following calculation formula to normalize the power data:

[0061]

[0062] Among them, X is the power data, X max and X min are the maximum and minimum values ​​in the power data, respectively, norm The data are normalized.

[0063] The simulation model construction module 30 is used to use sample training data and the physical characteristics of the power grid secondary equipment to capture the complex dynamic characteristics of the power grid secondary equipment through a deep neural network, thereby constructing a multi-level simulation model. Specifically, the simulation model construction module 30 may include a complex mathematical model construction unit 31, a multi-level simulation model construction unit 32, and an iterative optimization unit 33;

[0064] The complex mathematical model construction unit 31 is used to construct a complex network model of the following calculation formula through a deep neural network and a recursive neural network according to the sample training data and the physical characteristics of the power grid secondary equipment;

[0065] y t =σ(W h h t-1 +W x x t +b)

[0066] Among them, y t Represents the output at time t, x tCharacterize the input data, h t-1 Represents the hidden state of the previous moment, W h and W x is the weight matrix, b is the bias parameter, and σ(·) is the activation function;

[0067] The multi-level simulation model construction unit 32 is used to further construct a multi-level model based on a single deep neural network through the following calculation formula to capture the complex dynamic characteristics of the behavior of the secondary equipment of the power grid, and obtain a primary multi-level simulation model:

[0068]

[0069] Among them, h t (1) and h t (2) are the hidden states of the first and second layers respectively, W h (1) , W h (2) , W x (1) and is the weight matrix, b (1) , b (2) and b y is the bias parameter;

[0070] The iterative optimization unit 33 is used to iteratively optimize the primary multi-level simulation model based on the loss function to obtain the multi-level simulation model.

[0071] Further, the iterative optimization unit 33 is configured to perform the following operations when iteratively optimizing the primary multi-level simulation model based on the loss function:

[0072] S1: Initialize model parameters;

[0073] S2: In each iteration, the loss function is calculated based on the current model parameters;

[0074] S3: Calculate the gradient of the loss function and update the parameters according to the gradient descent method;

[0075] S4: Determine whether the updated parameters meet the convergence conditions; if so, stop the iteration and output the multi-level simulation model; otherwise, repeat steps S2-S4.

[0076] In this embodiment, first, the complex mathematical model construction unit 31 uses a deep neural network and a recursive neural network to construct a complex mathematical model based on sample training data and the physical characteristics of the power grid secondary equipment; then the multi-level simulation model construction unit 32 further constructs a multi-level model based on a single deep neural network, so as to better capture the complex dynamic characteristics of the device behavior. Finally, the loss function is solved by the gradient descent method, and the model converges by fast iterative calculation, so as to obtain an accurate and reliable multi-level simulation model.

[0077] For the iterative optimization unit 33, when performing iterative optimization calculation, the model parameters are first initialized. For example, the model parameter θ0 is initialized, the initial value is set, the model parameter η and the convergence threshold ε are initialized, etc. Then, in each iterative operation, the loss function L(θ) is calculated according to the current model parameter θ. The loss function can be obtained by the following calculation formula:

[0078]

[0079] Among them, y i is the actual output, is the predicted output, and N is the total number of samples.

[0080] Further, the iterative optimization unit 33 calculates the gradient of the loss function L(θ) And update the parameters according to the descending algorithm as shown below:

[0081]

[0082] Among them, θ k is the parameter value of the kth iteration, θ k+1 is the parameter value of the k+1th iteration, and η is the learning rate.

[0083] Further, the iterative optimization unit 33 performs convergence judgment. Specifically, the difference before and after the parameter update |θ is calculated k+1 -θ k |, if the difference is less than the set convergence threshold ε, that is, |θ k+1 -θ k |<ε, the iteration is stopped. If the difference does not reach the convergence threshold, the iteration operation continues and returns to the step loop of calculating the loss function.

[0084] After the multi-level simulation model is constructed, consider building a simulation platform for configuration. For example, in one embodiment, the simulation system may also include a simulation platform building module 60, which includes a hardware platform building unit 61 and a software platform building unit 62; the hardware platform building unit 61 is used to build a hardware platform using high-performance computing equipment or dedicated simulation hardware; the software platform building unit 62 is used to build a software platform that supports the integration and efficient calculation of complex models.

[0085] In this embodiment, the constructed multi-level simulation model is configured into the software platform, and the corresponding hardware platform is built accordingly. The hardware platform can use high-performance computing equipment or dedicated simulation hardware, and the software platform should support the integration of complex models and efficient computing performance. In this way, the simulation platform built based on the hardware platform and the software platform can be used to simulate the secondary equipment of the power grid.

[0086] The simulation execution module 40 is used to automatically optimize and adjust the corresponding key parameters to be simulated based on the multi-level simulation model according to the basic simulation parameters set by the user to obtain simulation result data. For example, in the simulation platform, the user sets the basic simulation parameters through a graphical interface, and the simulation platform automatically optimizes and adjusts the key parameters to improve the simulation performance and efficiency. When performing parameter optimization and adjustment, a particle swarm optimization algorithm can be used. Specifically, the simulation execution module 40 includes a parameter optimization adjustment unit, which is used to optimize and adjust the key parameters to be simulated through the following calculation formula:

[0087]

[0088] Among them, v i (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i at time t, p i is the individual optimal position of particle i, g is the global optimal position, ω is the inertia weight, c1 and c2 are acceleration constants, and r1 and r2 are random numbers.

[0089] The simulation result recording module 50 is used to monitor the changes of key parameters in real time during the simulation process, and to record and display the simulation result data. Specifically, in one embodiment, the simulation result recording module 50 may include a simulation parameter monitoring unit 51 and a simulation result display unit 52;

[0090] The simulation parameter monitoring unit 51 is used to monitor the changes of key parameters in real time during the simulation process and record the simulation result data in detail; wherein the simulation result data includes the response time, load condition and fault status of the power grid secondary equipment;

[0091] The simulation result display unit 52 is used to display the simulation result data through at least one of a chart, a report and a visualization tool.

[0092] The simulation result recording module 50 may further include an abnormal alarm unit 53 ; the abnormal alarm unit 53 is used to issue an alarm when the simulation result data monitored by the simulation parameter monitoring unit 51 is abnormal, so as to remind the staff to pay attention to the abnormality of the power grid secondary equipment.

[0093] In this embodiment, the simulation result recording module 50 can monitor the changes of key parameters in real time and record detailed simulation data, including device response time, load conditions and fault status, etc., and can also provide real-time feedback and warnings to ensure the stability and reliability of the simulation process.

[0094] Among them, the simulation system can simulate the device response time, that is, monitor the device's response time to the input signal, and evaluate the dynamic performance of the device. By continuously monitoring the response time, changes in the device response speed can be discovered in a timely manner to ensure the stability of the device under different working conditions.

[0095] The simulation system can also simulate load conditions, that is, the complex conditions of the equipment under different simulation conditions, and can analyze the performance of the equipment under different load conditions. The load condition data can be used to evaluate the equipment's carrying capacity and operating efficiency.

[0096] The simulation system can also simulate the fault state, that is, monitor the fault state of the equipment, record the time and type of the fault, analyze the cause and impact of the fault, and the real-time monitoring of the fault state helps to provide early warning and prevent the occurrence of serious faults. The key parameters monitored are further recorded in real time to generate detailed data logs, which can be used for subsequent analysis and optimization to help technicians understand the performance of the equipment under different simulation conditions. The recorded data may include timestamps, parameter values, and equipment status, etc., to ensure the integrity and accuracy of the data. In addition, the abnormal alarm unit 53 can also issue a warning immediately when an abnormality is detected to remind the staff. The alarm information may include equipment overload, response delay, fault warning, etc. The simulation system provides real-time feedback, displays the status and trend of the current simulation parameters, and helps the staff to adjust the simulation settings in time to ensure the stability and reliability of the simulation process.

[0097] In addition, the simulation system provided by the present invention further includes a simulation performance evaluation module 70;

[0098] The simulation performance evaluation module 70 is used to compare and analyze the actual test results of the power grid secondary equipment and the simulation result data through the following calculation formula to verify the reliability of the multi-level simulation model:

[0099]

[0100] Among them, R i is the model error about the ith simulation parameter, Y j and are the actual values ​​of the power grid secondary equipment and the model simulation predicted values, respectively, and M is the type of simulation parameter.

[0101] Meanwhile, the simulation system further comprises: a model updating module 80;

[0102] Model updating module 80, used to determine the error R of the i-th simulation parameter i Is it greater than the preset threshold? If it is less than, the multi-level simulation model is updated with the new power data until the error R of any simulation parameter is less than i are all smaller than the corresponding preset thresholds.

[0103] In this way, by verifying the accuracy and reliability of the simulation model, and when the accuracy and reliability of the model are difficult to meet the accuracy requirements, the model is updated with new sample data, thereby ensuring that the simulation model can continue to maintain high simulation accuracy and reliability. For reliability verification, the reliability of the model can be verified at preset intervals.

[0104] It can be seen that when this solution is implemented, detailed operating data is first obtained from the actual equipment and systems in the power grid through an advanced data acquisition system, which may include but is not limited to key parameters such as voltage, current, frequency, and power. These data are transmitted to the data processing module 20 for preprocessing. After the data processing is completed, a complex mathematical model is constructed based on the collected data and the physical characteristics of the power grid secondary equipment, and the deep neural network and the complex deep learning algorithm of the recursive neural network are used to describe the high-dimensional interaction between the device behavior and characteristics. At the same time, a high-performance power grid secondary equipment simulation platform is built, which includes a hardware platform and a software platform. The hardware platform can use high-performance computing equipment or dedicated simulation hardware. The software platform can include simulation calculation software and result visualization software. The software platform can support the integration and efficient calculation of multiple complex models. Then set the simulation parameters according to the simulation requirements, including operating conditions, fault conditions, and test schemes. The user interactively sets them through a graphical interface, and automatically adjusts the parameters through the optimization algorithm to optimize the simulation performance. The simulation platform performs simulation and uses the set parameters for efficient calculation. During the simulation process, the changes of key parameters are monitored in real time, and detailed simulation data, such as response time, load conditions, and fault status, are recorded. After that, the simulation results can be comprehensively analyzed to ensure the accuracy of the verification results, evaluate equipment performance, and diagnose faults. The analysis results are displayed through charts, reports, and visualization tools to facilitate user understanding and application. The generated report includes detailed simulation process, results, and recommended improvement measures. In the process of optimizing and adjusting the model and simulation parameters according to the simulation results, parameter adjustment, model improvement, and algorithm optimization can be included. Through the cyclic iterative process, the simulation accuracy and efficiency are gradually improved to ensure that the final results meet the actual application requirements. After the cyclic optimization is completed, the final selected feature subset and the optimized model are output. By comparing with the test results of the actual equipment, the accuracy and effectiveness of the model can be verified, providing support for the design, testing, and maintenance of power grid secondary equipment.

[0105] In summary, the power grid secondary equipment simulation system provided by the present invention has at least the following beneficial effects:

[0106] 1. In the present invention, by introducing complex mathematical models and control algorithms, high-precision simulation of power grid secondary equipment is achieved, and simulation efficiency and model adaptability are improved;

[0107] 2. In the present invention, through the advanced data acquisition system, a variety of key operating data can be obtained from the power grid in real time and accurately, providing a solid foundation for subsequent simulation analysis. The preliminary processing step ensures the accuracy and availability of the data and reduces the simulation errors caused by data quality problems;

[0108] 3. In the present invention, based on complex algorithms such as deep neural networks and recursive neural networks, the high-dimensional interactions and complex behavioral characteristics of power grid secondary devices can be accurately described. This advanced modeling method significantly improves the accuracy and reliability of simulation, making the simulation results closer to the actual operating conditions;

[0109] 4. In the present invention, the hardware and software platform built supports the integration of efficient calculations and complex models, ensuring the rapid response and accurate calculation of the simulation process. The graphical interface and automatic optimization algorithm enable users to easily set simulation parameters and quickly obtain simulation results, improving work efficiency. Simultaneously, the simulation results are displayed through charts, reports and visualization tools, making the analysis results more intuitive and easy to understand. This comprehensive analysis method helps users quickly understand the simulation results, evaluate equipment performance, and perform fault diagnosis;

[0110] 5. The present invention provides rich data support for subsequent simulation result analysis and model optimization by monitoring the changes of key parameters in real time and recording simulation data in detail during the simulation process. Such detailed data records help to deeply understand the device behavior and performance characteristics;

[0111] 6. The present invention optimizes and adjusts the model and simulation parameters through a cyclic iterative process to continuously improve the simulation accuracy and efficiency. The feature subsets and optimized models finally selected are verified by actual equipment tests to ensure the accuracy and effectiveness of the models. This continuous optimization mechanism enables the technical solution to continuously adapt to the new demands and challenges of power grid secondary equipment. The technical solution not only supports the design phase of power grid secondary equipment, but also provides strong support in the testing and maintenance phases. Through simulation analysis, the performance of the equipment under different working conditions can be predicted, providing a scientific basis for the design improvement of the equipment and the formulation of maintenance strategies.

[0112] The present specification also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute a method in any one of the embodiments in the specification.

[0113] The present specification also provides a computing device, including a memory and a processor, wherein executable codes are stored in the memory, and when the processor executes the executable codes, a method in any embodiment of the present specification is implemented.

[0114] The modules or units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. The above disclosure is only the preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiment are implemented, and the equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A power grid secondary equipment simulation system, characterized in that: include: Data acquisition module, data processing module, simulation model building module, simulation execution module and simulation result recording module; The data acquisition module is used to obtain power data of the secondary equipment of the power grid; wherein the power data at least includes current data and voltage data; The data processing module is used to perform data preprocessing on the power data to obtain sample training data; The simulation model building module is used to use the sample training data and the physical characteristics of the power grid secondary equipment to capture the complex dynamic characteristics of the power grid secondary equipment through a deep neural network to build a multi-level simulation model; The simulation execution module is used to automatically optimize and adjust the corresponding key parameters to be simulated based on the multi-level simulation model according to the simulation basic parameters set by the user to obtain simulation result data; The simulation result recording module is used to monitor the changes of key parameters in real time during the simulation process, and to record and display the simulation result data.

2. The power grid secondary equipment simulation system according to claim 1, characterized in that: The data processing module includes a normalization processing unit, which performs normalization processing on the power data using the following calculation formula: Among them, X is the power data, X max and X min are the maximum and minimum values ​​in the power data, respectively, norm The data are normalized.

3. The power grid secondary equipment simulation system according to claim 1, characterized in that: The simulation model construction module includes a complex mathematical model construction unit, a multi-level simulation model construction unit and an iterative optimization unit; The complex mathematical model construction unit is used to construct a complex network model of the following calculation formula through a deep neural network and a recursive neural network according to the sample training data and the physical characteristics of the power grid secondary equipment; y t =σ(W h h t-1 +W x x t +b) Among them, y t Represents the output at time t, x t Characterize the input data, h t-1 Represents the hidden state of the previous moment, W h and W x is the weight matrix, b is the bias parameter, and σ(·) is the activation function; The multi-level simulation model construction unit is used to further construct a multi-level model based on a single deep neural network through the following calculation formula to capture the complex dynamic characteristics of the behavior of the secondary equipment of the power grid, and obtain a primary multi-level simulation model: in, and are the hidden states of the first and second layers, respectively. and is the weight matrix, b (1) , b (2) and b y is the bias parameter; The iterative optimization unit is used to iteratively optimize the primary multi-level simulation model based on the loss function to obtain a multi-level simulation model.

4. The power grid secondary equipment simulation system according to claim 3, characterized in that: When the iterative optimization unit iteratively optimizes the primary multi-level simulation model based on the loss function, the iterative optimization unit is configured to perform the following operations: S1: Initialize model parameters; S2: In each iteration, the loss function is calculated based on the current model parameters; S3: Calculate the gradient of the loss function and update the parameters according to the gradient descent method; S4: Determine whether the updated parameters meet the convergence conditions; if so, stop the iteration and output the multi-level simulation model; otherwise, repeat steps S2-S4.

5. The power grid secondary equipment simulation system according to claim 1, characterized in that: The simulation system also includes a simulation platform building module, which includes a hardware platform building unit and a software platform building unit; The hardware platform building unit is used to build a hardware platform using high-performance computing equipment or dedicated simulation hardware; The software platform building unit is used to build a software platform that supports the integration and efficient calculation of complex models.

6. The power grid secondary equipment simulation system according to claim 1, characterized in that: The simulation execution module includes a parameter optimization and adjustment unit, which is used to optimize and adjust the key parameters to be simulated through the following calculation formula: Among them, v i (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i at time t, p i is the individual optimal position of particle i, g is the global optimal position, ω is the inertia weight, c1 and c2 are acceleration constants, and r1 and r2 are random numbers.

7. The power grid secondary equipment simulation system according to claim 1, characterized in that: The simulation result recording module includes a simulation parameter monitoring unit and a simulation result display unit; The simulation parameter monitoring unit is used to monitor the changes of key parameters in real time during the simulation process and record the simulation result data in detail; wherein the simulation result data includes the response time, load condition and fault status of the secondary equipment of the power grid; The simulation result display unit is used to display the simulation result data through at least one of a chart, a report and a visualization tool.

8. The power grid secondary equipment simulation system according to claim 7, characterized in that: The simulation result recording module also includes an abnormal alarm unit; The abnormal alarm unit is used to issue an alarm prompt when the simulation result data monitored by the simulation parameter monitoring unit is abnormal, so as to remind the staff to pay attention to the abnormality of the secondary equipment of the power grid.

9. The power grid secondary equipment simulation system according to claim 1, characterized in that: The simulation system also includes: a simulation performance evaluation module; The simulation performance evaluation module is used to compare and analyze the actual test results of the power grid secondary equipment and the simulation result data through the following calculation formula to verify the reliability of the multi-level simulation model: Among them, R i is the model error about the ith simulation parameter, Y j and are the actual values ​​of the power grid secondary equipment and the model simulation predicted values, respectively, and M is the type of simulation parameter.

10. The power grid secondary equipment simulation system according to claim 9, characterized in that: The simulation system also includes: a model updating module; The model updating module is used to determine the error R of the ith simulation parameter. i Is it greater than a preset threshold? If less than, the multi-level simulation model is updated with new power data until the error R of any simulation parameter is less than i are all smaller than the corresponding preset thresholds.

Images