Security event guarantee method and system for power grid operation

Abstract

The invention relates to the technical field of power grid monitoring, in particular to a security event guarantee method and system for power grid operation. Obtaining a long-range dependence factor according to the self-correlation attenuation characteristics of the monitoring data sequence; obtaining a periodic factor according to an energy intensity distribution characteristic in the spectrum data of the monitoring data sequence; performing time sequence decomposition on the monitoring data sequence to obtain a residual term; obtaining a trend factor according to fluctuation difference characteristics of the residual item and the monitoring data sequence; obtaining a nonlinear factor according to the nonlinear characteristic of the monitoring data sequence; obtaining fitness according to the long-range dependence factor, the periodic factor, the trend factor and the nonlinear factor; and obtaining a target monitoring data sequence according to the fitness of all the monitoring data sequences. According to the invention, prediction is carried out through the LSTM model according to the target monitoring data sequence, the fault risk in the power grid is analyzed according to the prediction result, and the accuracy of early warning of a security event and formulation of a recovery scheme after the event occurs is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid monitoring, and particularly to a safety event guarantee method and system for power grid operation. Background Art

[0002] In power grid operation, the timely and effective handling of safety events is the key to ensuring the stable operation of the power grid and reliable power supply. For example, safety events such as abnormal emergencies like a line being struck by lightning are generally divided into three important treatment measures. The first is event detection and preliminary response. The line protection device will quickly detect abnormal changes in electrical quantities such as current and voltage, and at the same time, the fault recorder will record the waveforms of electrical quantities before and after the fault, providing detailed data for subsequent fault analysis. The second is fault isolation and discharge protection, whose main function is to prevent the fault from further expanding. The system immediately activates the parallel discharge gap device. The parallel discharge gap is connected in parallel with the line insulator. When overvoltage appears on the line, the discharge gap breaks down and discharges, discharging overvoltage energy such as lightning current to the ground, reducing the risk of equipment damage, and buying time for subsequent fault handling. The third is the emergency command and recovery plan formulation of the control center. By collecting historical data provided by the SCADA (Supervisory Control And Data Acquisition) data acquisition and monitoring control system and the EMS (Energy Management System), the real-time operation state of the power grid is analyzed and predicted, and the impact of abnormal situations on the overall operation of the power grid is evaluated, such as changes in power flow distribution and voltage stability. This measure can combine with a machine learning prediction model, and according to the prediction results of the model, realize the analysis of the real-time operation state of the power grid and the assessment of the impact of faults, which has an important guaranteeing effect on the early warning of safety events and the formulation of recovery plans after the events occur.

[0003] SCADA and EMS are two important systems in circuit system automation, which can collect real-time data from various sensors, meters and devices, such as parameters like temperature, voltage, current and power. Such analyzed data are all time-series and have a fluctuating periodic trend. The traditional method believes that this type of data is suitable for processing using the common LSTM (Long Short-Term Memory) network model, which can effectively analyze time-series data and capture the changing law of power flow over time. However, due to the large number of time data indicators to be analyzed and processed, not all types of indicators are robust to the LSTM model. The reason may be related to the characteristics of the data itself and the degree of interference of historical data, resulting in relatively low prediction accuracy of some monitoring data, and ultimately affecting the accuracy of the early warning of safety events and the formulation of recovery plans after the events occur. Summary of the Invention

[0004] In order to solve the technical problem that the prediction accuracy of some monitoring data through the LSTM model is relatively low, which affects the early warning of safety events and the accuracy of the formulation of recovery plans after the events occur, the purpose of the present invention is to provide a safety event guarantee method and system for power grid operation, and the specific technical solutions adopted are as follows: Obtain different types of monitoring data sequences of each monitoring node in the power grid; Obtain a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence; obtain a period factor according to the energy intensity distribution characteristics in the spectrum data of the monitoring data sequence; perform time series decomposition on the monitoring data sequence to obtain a residual term; obtain a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence; obtain a non-linear factor according to the non-linear characteristics of the monitoring data sequence; Obtain a fitness according to the long-range dependence factor, the period factor, the trend factor and the non-linear factor; obtain a target monitoring data sequence according to the fitness of all monitoring data sequences; Perform prediction on the target monitoring data sequence through an LSTM model, and analyze the fault risk in the power grid according to the prediction result.

[0005] Further, the step of obtaining a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence includes: Remove the seasonal term of the monitoring data sequence through a time series decomposition algorithm to obtain a preprocessed sequence; calculate the autocorrelation coefficient of the preprocessed sequence at a preset initial lag order to obtain a first value; when the autocorrelation coefficient corresponding to any lag order infinitely approaches half of the first value, normalize the any lag order to obtain the long-range dependence factor.

[0006] Further, the step of obtaining a period factor according to the energy intensity distribution characteristics in the spectrum data of the monitoring data sequence includes: Perform discrete Fourier transform on the monitoring data sequence to obtain the spectrum data; calculate and normalize the ratio of the maximum value of the energy intensity in the spectrum data to the average value of the energy intensities of all frequencies to obtain the period factor.

[0007] Further, the step of obtaining a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence includes: Calculate and normalize the ratio of the standard deviation of the monitoring data sequence to the standard deviation of the residual term to obtain the trend factor.

[0008] Further, the step of obtaining a non-linear factor according to the non-linear characteristics of the monitoring data sequence includes: Calculate the Hurst exponent of the monitoring data sequence and perform a negative correlation mapping to obtain the non-linear factor.

[0009] Further, the step of obtaining a fitness according to the long-range dependence factor, the period factor, the trend factor and the non-linear factor includes: Calculate the average values of the long-range dependence factor, the periodic factor, the trend factor, and the non-linear factor to obtain the fitness of the monitoring data sequence.

[0010] Further, the step of obtaining the fitness according to the long-range dependence factor, the periodic factor, the trend factor, and the non-linear factor; and obtaining the target monitoring data sequence according to the fitness of all monitoring data sequences includes: Sort all the monitoring data sequences from largest to smallest according to the fitness, and use the monitoring data sequences before a preset position in the sorting as the target monitoring data sequences.

[0011] The present invention also provides a security event guarantee system for power grid operation, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps of any one of the security event guarantee methods for power grid operation.

[0012] The present invention has the following beneficial effects: In the present invention, the long-range dependence factor is obtained according to the autocorrelation decay characteristic of the monitoring data sequence, which can reflect whether the monitoring data sequence has a long-range dependence characteristic. The periodic factor is obtained according to the energy intensity distribution characteristic in the spectrum data of the monitoring data sequence, which can reflect whether the monitoring data sequence has a periodic characteristic. The trend factor is obtained according to the fluctuation difference characteristic between the residual term and the monitoring data sequence, which can reflect whether the monitoring data sequence has a trend characteristic. The non-linear factor is obtained according to the non-linear characteristic of the monitoring data sequence, which can reflect whether the monitoring data sequence has a non-linear change characteristic. Obtaining the fitness of the monitoring data sequence can comprehensively reflect whether it is suitable for prediction by the LSTM algorithm, improving the judgment accuracy. Obtaining the target monitoring data sequence from all the monitoring data sequences can determine the range suitable for prediction by the LSTM algorithm, eliminating the monitoring data sequences not suitable for this algorithm, making the credibility of the model prediction result higher. The power grid control center can analyze and evaluate the power grid operation state and the impact of faults more accurately based on the more accurate prediction result, improving the accuracy of early warning of security events and the formulation of recovery plans after the event occurs. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0014] Figure 1A flowchart of a method for ensuring security events in power grid operation provided by an embodiment of the present invention. Detailed implementation manners

[0015] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific implementation manners, structures, features and effects of a method and system for ensuring security events in power grid operation proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0017] The following specifically describes the specific solutions of a method and system for ensuring security events in power grid operation provided by the present invention with reference to the accompanying drawings.

[0018] Please refer to Figure 1 , which shows a flowchart of a method for ensuring security events in power grid operation provided by an embodiment of the present invention. The method includes the following steps: Step S1, obtaining different types of monitoring data sequences of each monitoring node in the power grid.

[0019] Obtaining different types of monitoring data sequences of each monitoring node in the power grid. In the embodiment of the present invention, the monitoring objects include the active power, reactive power, current, voltage amplitude and phase angle, and temperature of each line node, and the operating state parameters of the generators at the power generation end; the implementer can determine the monitoring objects according to the implementation scenario. Clean and preprocess the collected data. For the missing values in the data, use the interpolation method to fill them. For example, if the current data of a certain monitoring node is missing at a certain moment, linearly interpolate according to the data of the previous and subsequent moments to obtain an approximate value; perform maximum-minimum normalization processing on all the collected time series data to eliminate the influence of dimensions and facilitate subsequent analysis; after the data processing is completed, obtain different types of monitoring data sequences of each monitoring node. The acquisition range of the monitoring data sequences is the historical data within a period of time before the occurrence of a security accident, and the implementer can determine it according to the implementation scenario.

[0020] Step S2, obtaining a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence; obtaining a periodic factor according to the energy intensity distribution characteristics in the spectral data of the monitoring data sequence; performing time series decomposition on the monitoring data sequence to obtain a residual term; obtaining a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence; obtaining a non-linear factor according to the non-linear characteristics of the monitoring data sequence.

[0021] As a special type of recurrent neural network, the LSTM (Long Short-Term Memory) network can effectively process time-series data and capture the changing patterns of power flow distributions on different lines in current power grid security event analysis. By analyzing the principle of the LSTM algorithm, it can be found that this algorithm is more suitable for processing time-series data with long-range dependencies, strong periodicity, and trends. However, due to the characteristics of the data itself and the degree of interference in historical data, some types of monitored data sequences are not suitable for predictive analysis using this algorithm, resulting in low prediction accuracy and affecting the early warning of security events and the accuracy of formulating recovery plans after events occur. Therefore, it is necessary to determine the monitored data sequences with relatively high output result credibility when using the LSTM algorithm for prediction among all the monitored data sequences obtained.

[0022] Furthermore, traditional RNN (Recurrent Neural Network) has problems of gradient explosion and gradient vanishing when processing long sequences, resulting in the inability to effectively capture long-range dependency features. The LSTM, through its unique network design structure, designs a gating mechanism that can effectively solve the gradient problem and is more suitable for time-series data with long-range dependencies. Therefore, the long-range dependency factor can be obtained according to the autocorrelation decay characteristics of the monitored data sequence. Preferably, in the embodiments of the present invention, the steps of obtaining the long-range dependency factor include: removing the seasonal term of the monitored data sequence through a time-series decomposition algorithm to obtain a preprocessed sequence. It should be noted that the time-series decomposition algorithm belongs to the prior art, and the specific steps will not be elaborated. The preprocessed sequence after removing the seasonal term can avoid the influence of the periodic characteristics of the monitored data sequence on the acquisition of the long-range dependency factor and improve the accuracy of the long-range dependency factor. Calculate the autocorrelation coefficient of the preprocessed sequence at a preset initial lag order to obtain a first value. It should be noted that the calculation of the autocorrelation coefficient belongs to the prior art, and the specific steps will not be elaborated. The autocorrelation coefficient measures the correlation degree of the same event at different times. When the autocorrelation coefficient corresponding to any lag order infinitely approaches half of the first value, normalize the arbitrary lag order to obtain the long-range dependency factor, where the normalization is performed by normalizing the maximum and minimum values of all the arbitrary lag orders corresponding to the monitored data sequences. It should be noted that the preset initial lag order is 1. The larger the arbitrary lag order, the longer the half-life of the autocorrelation coefficient of the monitored data sequence, the slower the autocorrelation coefficient decreases as the lag order increases, the more obvious the long-range dependency characteristics of the monitored data sequence, the larger the long-range dependency factor, and the more suitable the monitored data sequence is for prediction using LSTM. The formula for obtaining the long-range dependency factor is as follows:

[0023] In the formula, represents the autocorrelation coefficient when the lag order is taken as x, represents the preset initial lag order, represents the first numerical value, U represents any lag order, represents normalization for any lag order, and C represents the long-range dependence factor.

[0024] LSTM can dynamically adjust the information weight through the gating mechanism, which can effectively adapt to the periodic patterns of time series data. At the same time, the cumulative characteristic of the memory unit enables it to capture the long-term trend of the data. Therefore, the LSTM algorithm is more suitable for the time series of monitoring data with periodic and trend characteristics, and the prediction accuracy is higher. Therefore, the period factor can be obtained according to the energy intensity distribution characteristics in the spectral data of the monitoring data sequence. Preferably, in the embodiment of the present invention, the steps of obtaining the period factor include: performing a discrete Fourier transform on the monitoring data sequence to obtain spectral data; it should be noted that the discrete Fourier transform belongs to the prior art, and the specific steps will not be elaborated. Each frequency in the spectral data corresponds to an energy intensity value. Calculate the ratio of the maximum value of the energy intensity in the spectral data to the average value of the energy intensities of all frequencies and normalize it to obtain the period factor; this normalization is the maximum-minimum normalization. When the period factor is larger, it means that the maximum value of the energy intensity in the spectral data is higher than the average level, and the peak feature is more obvious, then the periodic characteristic of the monitoring data sequence is more significant, and further the monitoring data sequence is more suitable for prediction using LSTM.

[0025] Furthermore, after obtaining the period factor of the monitoring data sequence, it is also necessary to determine whether there is a trend characteristic. Therefore, the monitoring data sequence is decomposed into time series to obtain the residual term; the trend factor is obtained according to the fluctuation difference characteristic between the residual term and the monitoring data sequence. Preferably, in the embodiment of the present invention, the steps of obtaining the trend factor include: calculating the ratio of the standard deviation of the monitoring data sequence to the standard deviation of the residual term and normalizing it to obtain the trend factor; this normalization is the maximum-minimum normalization. The reason for analyzing the decomposed residual term instead of the trend term in the process of calculating the trend factor is that: the decomposed trend part is usually a low-frequency smooth curve, usually obtained based on moving average or locally weighted regression, and its fluctuation amplitude may be artificially smoothed, making it difficult to directly reflect the true intensity of the trend; while the residual characteristic can reflect the quality of trend fitting. Since the residual term is the remaining part after subtracting the trend term and the seasonal term from the original monitoring data sequence, if the trend is the main component of the data, the residual fluctuation should be small, and if the trend is weak or does not exist, the residual fluctuation is more obvious; secondly, the residual term is not sensitive to outliers, which can reduce the influence of outliers, while directly analyzing the trend term is easily distorted by extreme values, resulting in low accuracy. If the trend characteristic of the monitoring data sequence is more obvious, the residual fluctuation characteristic is weaker. At this time, the standard deviation of the residual term will be much smaller than the standard deviation of the original monitoring data sequence, and the trend factor is larger, and the trend characteristic of the monitoring data sequence is more obvious, and further the monitoring data sequence is more suitable for prediction using LSTM.

[0026] The input gate, forget gate, and output gate of the LSTM all implement non-linear transformations through the sigmoid function and the tanh function, and can capture the complex non-linear dependence characteristics of time series data. The LSTM can dynamically adjust the non-linear influence through the gating mechanism, making the algorithm more adaptable to time series data with non-linear characteristic relationships in the monitored data sequence. Therefore, a non-linear factor is obtained according to the non-linear characteristics of the monitored data sequence; preferably, in the embodiment of the present invention, the steps of obtaining the non-linear factor include: calculating the Hurst exponent of the monitored data sequence and performing a negative correlation mapping to obtain the non-linear factor; it should be noted that the Hurst exponent belongs to the prior art and is usually calculated by the rescaled range analysis method, and the specific steps will not be elaborated. When the Hurst exponent is smaller, it means that the anti-persistence of the monitored data sequence is stronger, the non-linear characteristics are stronger, and it is more suitable for prediction by the LSTM algorithm.

[0027] Step S3, obtain the fitness according to the long-range dependence factor, periodic factor, trend factor, and non-linear factor; obtain the target monitored data sequence according to the fitness of all monitored data sequences.

[0028] After obtaining the long-range dependence factor, periodic factor, trend factor, and non-linear factor of the monitored sequence, the fitness can be obtained according to the long-range dependence factor, periodic factor, trend factor, and non-linear factor; preferably, in the embodiment of the present invention, the steps of obtaining the fitness include: calculating the average value of the long-range dependence factor, periodic factor, trend factor, and non-linear factor to obtain the fitness of the monitored data sequence. When the fitness is larger, it means that the monitored data sequence is suitable for prediction by the LSTM algorithm, and the accuracy of the prediction result is higher; when the fitness is smaller, it means that the accuracy of predicting the monitored data sequence using the LSTM algorithm is lower. Further, the monitored data sequence for LSTM prediction can be determined through the fitness, so the target monitored data sequence is obtained according to the fitness of all monitored data sequences; preferably, in the embodiment of the present invention, the steps of obtaining the target monitored data sequence include: sorting all monitored data sequences from largest to smallest according to the fitness, and taking the monitored data sequences before the preset position in the sorting as the target monitored data sequence. If the fitness of the overall monitored data sequence is larger, the preset position value can be adjusted accordingly to select more target monitored data sequences; if the fitness of the overall monitored data sequence is relatively low, the preset position value can be adjusted accordingly to select fewer target monitored data sequences to ensure the credibility of the prediction model result and improve the prediction efficiency at the same time; in the embodiment of the present invention, the preset position is the median of the sorting quantity; the target monitored data sequence can obtain a relatively accurate prediction result through prediction by the LSTM model.

[0029] Step S4, perform prediction through the LSTM model according to the target monitored data sequence, and analyze the fault risk in the power grid according to the prediction result.

[0030] After obtaining the target monitoring data sequence of all monitoring nodes in the power grid, the target monitoring data sequence can be predicted by LSTM, so as to analyze the operation state of the power grid and evaluate the impact of faults according to the prediction results. First, a training set and a test set of the target monitoring data sequence are constructed. The training set is used for parameter training of the model, and the test set is used for evaluating the performance of the model. During the training process, the hyperparameters of the LSTM model are adjusted to optimize the performance of the model, so that the prediction accuracy of the model meets the usage requirements. It should be noted that the LSTM model belongs to the prior art, and the specific training steps will not be elaborated. After training, by inputting the target monitoring data sequence for a period of time, the predicted value of the target monitoring data at the next moment can be obtained. After the model construction of the target monitoring data sequence is completed, the prediction and analysis of security events can be carried out. According to the prediction results of the model, the impact of faults can be evaluated. For example, if the prediction results of the model show that the active power of a certain key line has increased significantly and will exceed its rated capacity, and the power grid control center judges that its impact degree is relatively serious according to the prediction results, it means that there is an overload risk on this line, affecting the power supply reliability of the power grid. Furthermore, the control center needs to formulate a plan to ensure the stable operation of this line and reduce the impact of faults on the power grid. Thus, by analyzing whether each monitoring node is suitable for the LSTM model, the appropriate data type is selected for prediction, improving the prediction accuracy of the monitoring data in the power grid, avoiding the errors of prediction results that are not suitable for this model for prediction, and finally improving the accuracy of early warning of security events and the formulation of recovery plans after the events occur.

[0031] In summary, the embodiment of the present invention provides a method for ensuring security events in power grid operation; obtaining a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence; obtaining a period factor according to the energy intensity distribution characteristics in the spectral data of the monitoring data sequence; performing time series decomposition on the monitoring data sequence to obtain a residual term; obtaining a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence; obtaining a non-linear factor according to the non-linear characteristics of the monitoring data sequence; obtaining a fitness according to the long-range dependence factor, the period factor, the trend factor and the non-linear factor; obtaining a target monitoring data sequence according to the fitness of all monitoring data sequences. The present invention predicts according to the target monitoring data sequence through the LSTM model, analyzes the fault risks in the power grid according to the prediction results, and improves the accuracy of early warning of security events and the formulation of recovery plans after the events occur.

[0032] The present invention also proposes a security event guarantee system for power grid operation, including a memory, a processor, and a computer program stored in the memory and operable on the processor, and the processor executes the computer program to implement the steps of any one of the methods for ensuring security events in power grid operation.

[0033] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0034] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A safety event guarantee method for power grid operation, characterized in that, The method includes the following steps: Obtain different types of monitoring data sequences of each monitoring node in the power grid; Obtain a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence; obtain a period factor according to the energy intensity distribution characteristics in the spectrum data of the monitoring data sequence; perform time series decomposition on the monitoring data sequence to obtain a residual term; obtain a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence; obtain a non-linear factor according to the non-linear characteristics of the monitoring data sequence; Obtain a fitness according to the long-range dependence factor, the period factor, the trend factor, and the non-linear factor; obtain a target monitoring data sequence according to the fitness of all monitoring data sequences; Perform prediction on the target monitoring data sequence through an LSTM model, and analyze the fault risk in the power grid according to the prediction result.

2. The safety event guarantee method for power grid operation according to claim 1, characterized in that The step of obtaining a long-range dependence factor according to the autocorrelation decay characteristics of the monitoring data sequence includes: Remove the seasonal term of the monitoring data sequence through a time series decomposition algorithm to obtain a preprocessed sequence; calculate the autocorrelation coefficient of the preprocessed sequence at a preset initial lag order to obtain a first value; when the autocorrelation coefficient corresponding to any lag order infinitely approaches half of the first value, normalize the any lag order to obtain the long-range dependence factor.

3. The safety event guarantee method for power grid operation according to claim 1, wherein The step of obtaining a period factor according to the energy intensity distribution characteristics in the spectrum data of the monitoring data sequence includes: Perform a discrete Fourier transform on the monitoring data sequence to obtain the spectrum data; calculate and normalize the ratio of the maximum value of the energy intensity in the spectrum data to the average value of the energy intensities of all frequencies to obtain the period factor.

4. A method for ensuring safety events in power grid operation according to claim 1, characterized in that, The step of obtaining a trend factor according to the fluctuation difference characteristics between the residual term and the monitoring data sequence includes: Calculate and normalize the ratio of the standard deviation of the monitoring data sequence to the standard deviation of the residual term to obtain the trend factor.

5. A method for ensuring safety events in power grid operation according to claim 1, characterized in that, The step of obtaining a non-linear factor according to the non-linear characteristics of the monitoring data sequence includes: Calculate the Hurst exponent of the monitoring data sequence and perform a negative correlation mapping to obtain the non-linear factor.

6. A safety event guarantee method for power grid operation according to claim 1, characterized in that, The step of obtaining a fitness according to the long-range dependence factor, the period factor, the trend factor, and the non-linear factor includes: Calculate the average value of the long-range dependence factor, the period factor, the trend factor, and the non-linear factor to obtain the fitness of the monitoring data sequence.

7. A safety event guarantee method for power grid operation according to claim 1, characterized in that, Obtain a fitness according to the long-range dependence factor, the period factor, the trend factor, and the non-linear factor; The step of obtaining a target monitoring data sequence according to the fitness of all monitoring data sequences includes: Sort all monitoring data sequences from large to small according to the fitness, and use the monitoring data sequences before a preset position in the sorting as the target monitoring data sequences.

8. A security event guarantee system for power grid operation, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-7.

Images