An energy twin-based power grid fault early warning system

Abstract

The application discloses a kind of power grid fault early warning system based on energy twin body;Including physical system, data transmission layer and digital twin system;The digital twin system is according to the mobile data that power equipment abnormal data is synchronized into digital twin database to promote twin body model real-time synchronous update, and carry out each scene simulation deduction of twin system;The present application is by constructing distributed energy twin body and source network load storage collaborative optimization operation deduction, timely find the problem such as actual power grid overload, voltage overline, early warning to possible fault;Combined with physical system, data transmission layer and digital twin technology, through real-time data acquisition, accurate modeling and intelligent analysis, realize real-time monitoring, fault early warning and collaborative optimization to power grid, with strong intelligentization and self-adaptive ability, it has wide application prospect in power grid management and dispatching, especially in the security and stability of power supply.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power grid fault early warning, more specifically, especially relates to a power grid fault early warning system based on energy twins. BACKGROUND

[0002] As a highly complex system, the power grid covers multiple links from power generation, transmission to distribution. With the change of energy structure, especially the large-scale access of renewable energy (such as wind power, photovoltaic, etc.), the operating environment of the power grid has become more complex. The power grid needs to cope with dynamic changes in load, external weather factors and the state of equipment, which makes the occurrence of power grid faults more unpredictable. Power grid faults usually manifest in the following types:

[0003] Short circuit fault: such as single-phase grounding, three-phase short circuit, etc., often occurs in substation, transmission line and other equipment.

[0004] Overload fault: the load of the power grid exceeds the design range, causing equipment damage or power supply interruption.

[0005] Equipment failure: due to equipment aging or failure, leading to partial failure of the power grid.

[0006] Power quality problems: such as voltage fluctuation, frequency instability, etc.

[0007] With the continuous development of the power industry and the promotion of digital transformation, digital twin technology provides a new solution to the various challenges faced by the power industry. Based on the digital twin model mapping the physical architecture of the power system, the power equipment is digitally modeled, facilitating power equipment management and monitoring; at the same time, based on the power system, data is obtained from each power system operation module, a virtual model is established to simulate the physical process in the real world, the power grid operation is monitored in real time, and data is collected, analyzed and optimized, realizing the flow of data in the digital twin model, forming a physical entity and a three-dimensional model that ensure consistent behavior through data, showing a virtual-real interactive synchronous state, and the instructions issued through the digital twin model control the physical equipment, achieving the effect of virtual control of real, providing a new technical framework and solution for the collaborative interaction between source, grid, load and storage.

[0008] Digital twin refers to the establishment of a virtual model of a physical entity (such as power grid equipment, system, etc.), which interacts with the physical entity in real time to simulate and predict the behavior of the physical system. In the field of power grid, digital twin technology can be applied to equipment state monitoring, fault prediction and diagnosis, intelligent dispatching, etc.

[0009] However, the prior art has some problems: with the application of digital twin technology in the power system, a large amount of data is collected and processed, but data islands are caused, synchronous transmission of data cannot be realized, rapid processing of data cannot be realized, and real-time data-based power grid fault early warning cannot be realized, so we propose a power grid fault early warning system based on energy twin. SUMMARY

[0010] In view of the problems in the prior art, the purpose of the present application is to provide a power grid fault early warning system based on energy twin, which carries out regional source network load storage resource panoramic monitoring through a deduction method based on digital twin technology, realizes dynamic updating of power grid topology and dynamic monitoring of equipment state, and establishes power grid twins under each voltage level by mirroring physical objects of the power system in virtual space through digital twin technology and correcting and updating the digital twin power grid and the regulation and control physical power grid through virtual-real interaction capability.

[0011] To achieve the above purpose, the present application provides the following technical scheme: a power grid fault early warning system based on energy twin, comprising a physical system, a data transmission layer and a digital twin system.

[0012] The physical system is used for collecting data, and the physical system obtains the addition and replacement information of power equipment according to the input data, and also obtains daily type data and environmental data through sensors, wherein the daily type data includes maintenance data and fault data of power equipment, and the physical system retrieves photovoltaic data, wind power data and load history data of distributed new energy stored in the database, and the physical system transmits the collected data to the digital twin system through the data transmission layer.

[0013] The data transmission layer synchronously transmits the collected data to the digital twin system, the digital twin system stores the collected data into a digital twin database, the digital twin system pre-processes and analyzes the collected data, and then calculates and processes the photovoltaic data, the wind power data and the load history data to generate photovoltaic models and wind power models.

[0014] The calculation formula of the photovoltaic model is as follows:

[0015]

[0016] Among them, Pmax represents the maximum power output of the photovoltaic output, Pv represents the power output of each photovoltaic panel, ψ represents the conversion efficiency of the photovoltaic panel, and a represents the area of the photovoltaic panel. pv represents the temperature coefficient, T represents the battery temperature, and T ref represents the set battery temperature, and S represents the area of the photovoltaic panel.

[0017] The calculation formula of the wind power model is as follows:

[0018]

[0019] wherein, represents the maximum power output of the wind power, and V w represents the actual wind speed, represents the wind speed input of the wind power, represents the wind speed output of the wind power, represents the reference wind speed of the wind power, Pv represents the power output of each wind power, and ζ represents the conversion efficiency of the wind power.

[0020] The digital twin system synchronizes the mobile data in the digital twin database according to the power equipment transaction data, promotes real-time synchronization update of the twin model, and carries out scene simulation deduction of the twin system. The digital twin system simulates power demand and supply under different scenarios to optimize and deduce the power grid.

[0021] Specifically, the data transmission layer includes switches, servers and firewalls. The switches are used to realize device exchange between the physical system and the digital twin system, and are mainly used to forward data frames from the physical system to the digital twin system. The servers are used to realize the establishment of communication connection between the physical system and the digital twin system. The firewalls are used to realize the security protection of the transmitted data, and ensure the stability of data transmission.

[0022] Specifically, the synchronization transmission of the data transmission layer is to share a common clock signal between the physical system and the digital twin system. The clock information is transmitted through physical connection or control signal. The physical system provides the clock signal while transmitting data, and the digital twin system receives data at appropriate time according to the clock signal. Alternatively, the physical system and the digital twin system realize synchronization through the synchronization word in the data frame. The digital twin system determines the transmission time sequence of data by analyzing the synchronization word or other control information in the frame header.

[0023] Specifically, the preprocessing is used for noise removal, abnormal value detection, missing value detection, missing value filling and normalization processing.

[0024] The removing noise is used for filtering noise in the data, improving the accuracy of the data, the detecting outliers is used for detecting abnormal data in the data, and the abnormal data is removed, the detecting missing values is used for detecting missing values in the data, the filling missing values is used for mean filling the positions of the removed abnormal values and the positions of the missing values in the data, and the normalization processing is used for scaling the data between [0, 1].

[0025] Specifically, the removing noise is used for smoothing and removing noise of the data, and the removing noise uses the following calculation formula to process the data:

[0026]

[0027] Wherein, X t represents the original data corresponding to the time point t in the data, S t-1 represents the smoothing value corresponding to the time point t-1 in the data, and a is a smoothing factor, which is used to allocate the weights of the new data point and the historical data point, and W f (a,b) represents the transformation structure of the data in scale a and position b, ψ * is the conjugate of wavelet transform, a is a scale parameter, and b is a translation parameter.

[0028] Specifically, the detecting outliers detects through the following algorithm:

[0029]

[0030] Wherein, Z represents the data cleaning calculation value, X represents the data, u represents the mean of the data group in the data, σ represents the standard deviation of the data group in the data, n represents the number of data in the data group, and i represents the position of the data in the data group, x i is the ith data in the data group, and for the abnormal judgment, |Z|>2, the data point is an outlier, and the outlier is removed.

[0031] Specifically, the normalization processing maps the data between [0, 1], and the normalization formula is as follows:

[0032] Wherein, X min represents the minimum value in the data, X max represents the maximum value in the data.

[0033] The filling missing values uses average number to replace processing:

[0034] Wherein, m is the amount of non-anomalous data.

[0035] Specifically, the analysis processing is used to extract features in the data, and the extracted features are as follows:

[0036] The calculation of the mean value is as follows:

[0037]

[0038] Where μ old is the mean value of the previous data, μ new is the mean value of the new data, and N is the number of data.

[0039] And the calculation formula of the mean value is as follows:

[0040]

[0041] Where μ is the calculated mean value, w i is the weight value of the data, and x i is the data value.

[0042] The calculation of the standard deviation is as follows:

[0043]

[0044] Where μ is the mean value, x i is the data value, N is the number of data, and σ is the calculated standard deviation.

[0045] The calculation of the variance is as follows:

[0046]

[0047] Where μ is the mean value, x i is the data value, N is the number of data, and σ 2 is the calculated variance.

[0048] The calculation of the maximum and minimum values is as follows:

[0049] Max new = max(Max old , x new )

[0050] Min new = min(Min old , x new ),

[0051] Where Max new is the new maximum value after calculation, Max old is the previous maximum value, Min new is the new minimum value after calculation, Min old is the previous minimum value, and x new is the newly added data.

[0052] Specifically, the digital twin system fuses the mean X1, the standard deviation X2, the variance X3, the maximum value X4 and the minimum value X5 through feature cross technology;

[0053]

[0054] Obtain a new feature of the digital twin system through feature fusion;

[0055] The fault prediction is carried out through a multilayer perception machine, and the calculation formula of the multilayer perception machine is:

[0056]

[0057] Wherein, W1, W2…W k is the weight matrix of the full connection layer, b1, b2…b k is the bias term, ReLU is the activation function, h1, h2…h k is the intermediate representation of each layer, and the cross feature X cross is further learned through the full connection layer, and finally the prediction result y is output.

[0058] Specifically, the digital twin model simulates on the basis of real-time data flow, adjusts the behavior of the physical system through model predictive control, and the calculation formula of the model predictive control is as follows:

[0059]

[0060] Wherein, u is the optimal control input, y k is the system output, y ref,k is the reference trajectory, Q and R are weight matrices, and N is the prediction time domain length.

[0061] The technical effects and advantages of the present application are:

[0062] The present application realizes the construction of distributed energy twin body and source network load storage collaborative optimization operation deduction, timely discovers the problems of actual power grid such as heavy overload and voltage overline, prewarns the possible faults, improves the scheduling decision-making ability under multiple scenes, and reduces the potential operation risk;

[0063] The present application fully utilizes the digital twin technology, realizes the comprehensive collection and comprehensive analysis of power grid operation state data, realizes the construction of distributed energy twin body, operation deduction scene arrangement and multi-state deduction simulation, timely discovers the weak points of actual power grid, prewarns the possible faults, presets the faults that have not occurred, verifies the feasibility and usability of scheduling optimization strategy, effectively promotes the coordinated operation of multiple distributed devices and power systems, significantly improves the scheduling decision-making ability under multiple scenes, and reduces the potential operation risk;

[0064] Combined with physical systems, data transmission layers, and digital twin technologies, real-time monitoring, fault early warning, and collaborative optimization of power grids are achieved through real-time data collection, accurate modeling, and intelligent analysis, with strong intelligent and adaptive capabilities, and wide application prospects in power grid management and dispatch, especially in ensuring the safety and stability of power supply.

[0065] Real-time fault early warning: The digital twin system can simulate and predict fault data of power grid equipment to provide early warning before faults occur, reducing sudden power outages and equipment damage.

[0066] Power grid optimization: By simulating different load and power supply scenarios, the system can achieve collaborative optimization of power grid operation, ensuring energy supply and demand balance and improving power grid operation efficiency.

[0067] Data-driven decision support: With big data analysis, machine learning, and artificial intelligence technologies, the system can continuously learn and adjust prediction models based on historical data to improve prediction accuracy.

[0068] Other features and advantages of the present application will become apparent from the following detailed description of exemplary embodiments thereof, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0069] Fig. 1 is a system structure schematic diagram provided by the present application;

[0070] Fig. 2 is a flowchart provided by the present application. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0072] As shown in Figs. 1-2 , the power grid fault early warning system based on energy twin provided by the embodiments of the present application includes a physical system, a data transmission layer, and a digital twin system.

[0073] The physical system is used to collect data. It acquires information on the addition and replacement of power equipment based on the entered data. The physical system also acquires daily data and environmental data through sensors. The daily data includes maintenance data and fault data of power equipment. Furthermore, the physical system retrieves photovoltaic data, wind power data, and historical load data of distributed new energy sources stored in the database. The physical system transmits the collected data to the digital twin system through the data transmission layer.

[0074] The data transmission layer synchronously transmits the collected data to the digital twin system. The digital twin system stores the collected data in the digital twin database. The digital twin system preprocesses and analyzes the collected data, and then performs calculations on the photovoltaic data, the wind power data, and the historical load data to generate photovoltaic models and wind power models.

[0075] The calculation formula for the photovoltaic model is as follows:

[0076]

[0077] in, This represents the maximum power output of the photovoltaic system. The power output of each photovoltaic panel, ψ represents the conversion efficiency of the photovoltaic panel, and α pv It is expressed as a temperature coefficient, where T represents the battery temperature. ref This represents the set battery temperature, and S represents the area of ​​the photovoltaic panel.

[0078] The calculation formula for the wind power model is as follows:

[0079]

[0080] in, V represents the maximum output power of the wind power. w Expressed as actual wind speed. This represents the wind speed at which wind power is input. This represents the wind speed output by the wind power generator. This represents the base wind speed for wind power. The power output of each wind turbine, ζ represents the wind power conversion efficiency;

[0081] The digital twin system uses mobile data synchronized to the digital twin database based on power equipment anomaly data to drive real-time synchronous updates of the twin model and conducts simulations of various scenarios of the twin system. The digital twin system performs collaborative optimization simulations of the power grid by simulating power demand and supply under different scenarios.

[0082] In this embodiment, preferably, the data transmission layer includes a switch, a server and a firewall, the switch is used to realize device exchange between the physical system and the digital twin system, mainly for forwarding data frames from the physical system to the digital twin system, the server is used to realize the establishment of communication connection between the physical system and the digital twin system, and the firewall is used to realize the security protection of the transmitted data, ensuring the stability of data transmission.

[0083] It should be noted that the design of the data transmission layer ensures that the data exchange between the physical system and the digital twin system is not only fast and efficient, but also ensures the security and stability during transmission, which can support the efficient operation of the power grid fault early warning system, especially when facing massive data and potential security risks, providing strong protection.

[0084] In this embodiment, preferably, the synchronous transmission of the data transmission layer is that the physical system and the digital twin system share a common clock signal, the clock information is transmitted through physical connection or control signal, the physical system provides clock signal while transmitting data, and the digital twin system receives data at appropriate time according to the clock signal, or the physical system and the digital twin system realize synchronization through synchronization words in data frames, and the digital twin system determines the transmission timing of data by analyzing the synchronization words or other control information in the frame header.

[0085] It should be noted that the two ways of sharing common clock signal or synchronization word are used to realize efficient and accurate synchronization between the physical system and the digital twin system, the shared clock signal is suitable for scenarios with extremely strict timing requirements, while the synchronization word method provides a more flexible and scalable solution, both have their own advantages, and the appropriate synchronization method can be selected according to the actual application requirements to ensure the efficiency of data transmission and the accuracy of timing.

[0086] In this embodiment, preferably, the preprocessing is used for removing noise, detecting abnormal values, detecting missing values, filling missing values and normalization processing.

[0087] The noise removal is used to filter the noise in the data to improve the accuracy of the data, the abnormal value detection is used to detect abnormal data in the data and eliminate abnormal data, the missing value detection is used to detect missing values in the data, the missing value filling is used to fill the positions of the eliminated abnormal values and the missing values in the data with mean value, and the normalization processing is used to scale the data to [0, 1].

[0088] It should be noted that by preprocessing the data, the quality of the data is significantly improved, the noise is removed, the outliers are detected and removed, the missing values are detected and filled, and the normalization processing can ensure the accuracy, integrity and consistency of the data, lay a solid foundation for subsequent data analysis, modeling and prediction, and improve the performance of the model and the reliability of the results.

[0089] In this embodiment, preferably, the noise removal is used for smoothing and removing noise of the data, and the noise removal is processed by using the following calculation formula:

[0090]

[0091] Wherein, X t represents the original data corresponding to the time point t in the data, S t-1 represents the smoothing value corresponding to the time point t-1 in the data, and a is a smoothing factor for allocating the weights of the new data point and the historical data point, and W f (a,b) represents the transformation structure of the data in scale a and position b, ψ * is the conjugate of wavelet transform, a is a scale parameter, and b is a translation parameter.

[0092] It should be noted that by using wavelet transform combined with smoothing factor for noise removal, not only the noise in the data can be effectively removed, but also the key features of the data can be preserved. Wavelet transform is a method that can analyze signals in multiple scales and multiple resolutions, which can effectively capture local features in signals. At the same time, smoothing processing can remove high-frequency noise, thereby improving the quality of the data.

[0093] In this embodiment, preferably, the detection of outliers is detected by the following algorithm:

[0094]

[0095] Wherein, Z represents the data cleaning calculation value, X represents the data, u represents the mean of the data group in the data, σ represents the standard deviation of the data group in the data, n represents the number of data in the data group, and i represents the position of the data in the data group, x i is the ith data in the data group, and for the determination of the abnormal value, |Z|>2, the data point is an outlier, and the outlier is removed.

[0096] It should be noted that by calculating the mean, standard deviation and other related statistics of the data group, the extreme outliers in the data can be effectively removed.

[0097] In this embodiment, preferably, the normalization processing maps the data between [0, 1], and the normalization formula is as follows:

[0098] wherein X min represents the minimum value in the data, X max represents the maximum value in the data;

[0099] The missing values are replaced by the average number:

[0100] wherein m is the amount of non-anomalous data;

[0101] It should be noted that the data is scaled by the normalization process, so that the data is mapped between [0, 1], which can reduce the dimension of the data, facilitate subsequent data calculation, improve the efficiency of the calculation and reduce the complexity of the calculation, and the missing values are filled by the mean value, which can improve the completeness of the data.

[0102] In this embodiment, preferably, the analysis process is used to extract features in the data, and the extracted features are as follows:

[0103] The mean value is calculated as follows:

[0104]

[0105] wherein μ old is the mean value of the previous data, μ new is the mean value of the new data, and N is the number of data;

[0106] And the mean value is calculated as follows:

[0107]

[0108] wherein μ is the calculated mean value, w i is the weight value of the data, and x i is the data value;

[0109] The standard deviation is calculated as follows:

[0110]

[0111] wherein μ is the mean value, x i is the data value, N is the number of data, and σ is the calculated standard deviation;

[0112] The variance is calculated as follows:

[0113]

[0114] wherein μ is the mean value, x i is the data value, N is the number of data, and σ 2 is the calculated variance;

[0115] The maximum and minimum are calculated as follows:

[0116] Max new = max(Max old , x new )

[0117] Min new = min(Min old , x new ),

[0118] where Max new is the new maximum after calculation, Max old is the previous maximum, Min new is the new minimum after calculation, Min old is the previous minimum, and x new is the newly added data.

[0119] It should be noted that.

[0120] In this embodiment, preferably, the digital twin system fuses the mean X1, the standard deviation X2, the variance X3, the maximum value X4 and the minimum value X5 through feature cross technology.

[0121]

[0122] to obtain a new feature of the digital twin system.

[0123] The multilayer perceptron is used for fault prediction, and the calculation formula of the multilayer perceptron is:

[0124]

[0125] where W1, W2…W k is the weight matrix of the full connection layer, b1, b2…b k is the bias term, ReLU is the activation function, h1, h2…h k is the intermediate representation of each layer, and cross feature X cross is further learned through the full connection layer, and finally outputs the prediction result y.

[0126] It should be noted that feature cross is to fuse and combine the mean, standard deviation, variance, maximum value and minimum value in the data as a new feature to capture the interaction between them, which helps the linear model or other traditional models to better model the complex nonlinear relationship, and then through the multilayer perceptron to realize the prediction of the fault information, and obtain the information of the power grid fault.

[0127] In this embodiment, preferably, the digital twin model is simulated based on real-time data streams, and the behavior of the physical system is adjusted through model predictive control. The calculation formula for model predictive control is as follows:

[0128]

[0129] Where u is the optimal control input, y k This is the system output, y ref,k The reference trajectory is Q and R, which are weight matrices, and N is the prediction time domain length.

[0130] It should be noted that Model Predictive Control (MPC) is used to optimize and adjust the behavior of a physical system. Based on the dynamic model of the system, MPC continuously optimizes and calculates, and adjusts the control input in real time so that the system output tracks the reference trajectory as closely as possible, ensuring that the system behavior is consistent with the target reference trajectory, thereby achieving efficient system control and regulation.

[0131] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An energy twin based power grid failure warning system, characterized in that, The physical system, the data transmission layer and the digital twin system are included; The physical system is used for collecting data, and the physical system obtains the addition and replacement information of power equipment according to the input data, obtains daily type data and environment data through sensors, the daily type data includes maintenance data and fault data of power equipment, and the physical system retrieves photovoltaic data, wind power data and load history data of distributed new energy stored in the database, and the physical system transmits the collected data to the digital twin system through the data transmission layer; The data transmission layer synchronously transmits the collected data to the digital twin system, the digital twin system stores the collected data into the digital twin database, the digital twin system pre-processes and analyzes the collected data, then calculates and processes the photovoltaic data, the wind power data and the load history data to generate photovoltaic model and wind power model; The calculation formula of the photovoltaic model is as follows: ， wherein, the maximum power expressed as photovoltaic output, the power output by each photovoltaic panel, the conversion efficiency of the photovoltaic panel, expressed as a temperature coefficient, expressed as a cell temperature, expressed as a set cell temperature, expressed as an area of the photovoltaic panel; The calculation formula of the wind power model is as follows: ， wherein, represents the maximum power of the wind power output, represents the actual wind speed, represents the wind speed of the wind power input, represents the wind speed of the wind power output, represents the reference wind speed of the wind power, the power of each wind power output, represents the conversion efficiency of the wind power; The digital twin system synchronously updates the twin model in real time according to the moving data of the power equipment transaction data in the digital twin database, and develops scene simulation deduction of the digital twin system, and the digital twin system optimizes and deduces the power grid by simulating power demand and supply under different scenarios.

2. The energy twin based power grid failure warning system as claimed in claim 1, wherein: The data transmission layer includes switches, servers and firewalls, the switches are used to realize equipment exchange between the physical system and the digital twin system, mainly used for forwarding data frames from the physical system to the digital twin system, the servers are used to realize communication connection between the physical system and the digital twin system, and the firewalls are used to realize security protection of the transmitted data, and ensure stable data transmission.

3. The energy twin based power grid failure warning system as claimed in claim 1, wherein: The synchronous transmission of the data transmission layer is that the physical system and the digital twin system share a common clock signal, and the clock information is transmitted through physical connection or control signal, the physical system provides clock signal while transmitting data, and the digital twin system receives data at appropriate time according to the clock signal, or the physical system and the digital twin system realize synchronization through synchronization word in data frame, and the digital twin system determines the transmission time sequence of data by analyzing the synchronization word or other control information in the frame header.

4. The energy twin based power grid failure warning system of claim 1, wherein: The preprocessing is used for removing noise, detecting abnormal values, detecting missing values, filling missing values and normalizing data; The noise removal is used for filtering noise in data to improve data accuracy, the abnormal value detection is used for detecting abnormal data in data and eliminating abnormal data, the missing value detection is used for detecting missing values in data, the missing value filling is used for mean filling in the positions of eliminated abnormal values and missing values in data, and the normalization processing is used for scaling data to [0, 1].

5. A power grid failure warning system based on energy twins according to claim 4, characterized in that: The noise removal is used for smoothing and removing noise of the data, and the noise removal processes the data by using the following calculation formula: ， wherein, denotes the original data at time point t in the data, denotes the original data at time point the corresponding smoothed value, is a smoothing factor used to assign weights to the new data point and the historical data points, and denotes the transformation structure of the data in scale and position , is the conjugate of the wavelet transform, is a scale parameter, is a translation parameter.

6. The energy twin based power grid failure warning system of claim 4, wherein: The detection of the abnormal value is detected by the following algorithm: ， wherein, is calculated as a data cleaning value, is data, is calculated as a mean of data groups in data, is calculated as a standard deviation of data groups in data, and is calculated as a number of data in data groups, and is calculated as a position of data in data groups, is the first data in data groups, and for the determination of an anomaly > 2, the data point is an outlier and the outlier is rejected.

7. The energy twin based power grid failure warning system of claim 4, wherein: The normalization processing maps the data between [0, 1], and the normalization formula is as follows: wherein, denotes the minimum value in the data, denotes the maximum value in the data; The missing value is replaced by the average value: wherein, is a non-anomalous data volume.

8. The energy twin based power grid failure warning system of claim 1, wherein: The analysis processing is used for extracting the features in the data, and the extracted features are as follows: The calculation of the mean value is as follows: ， wherein, is the mean of previous data, is the mean of new data, is the number of data; And the calculation formula of the mean value is as follows: ， wherein, is the calculated mean, is the weight value of the data, is the data value; The calculation of the standard deviation is as follows: ， wherein is the mean, is the data value, is the number of data, is the standard deviation calculated; The calculation of the variance is as follows: ， wherein, is the mean, is the data value, is the number of data, is the calculated variance; The calculation of the maximum and minimum values is as follows: ， wherein, is the new maximum value after the calculation, is the previous maximum value, is the new minimum value after the calculation, is the previous minimum value, is the newly added data.

9. The energy twin based power grid failure warning system of claim 1, wherein: The digital twin system fuses the mean value , the standard deviation , the variance , the maximum value and the minimum value by feature crossing techniques; ， Get the feature fusion new feature of the digital twin system; The fault prediction is performed by the multilayer perception machine, and the calculation formula of the multilayer perception machine is: ， wherein, is a weight matrix of the fully connected layer, is a bias term, is an activation function, is an intermediate representation of each layer, cross features further learning through the fully connected layer, finally output the prediction result .

10. The energy twin based power grid failure warning system of claim 1, wherein: The digital twin system simulates on the basis of real-time data flow, adjusts the behavior of the physical system through model predictive control, and the calculation formula of the model predictive control is as follows: ， wherein, is the optimal control input, is the system output, is the reference trajectory, is the weight matrix, is the prediction horizon length.

Images