Power system voltage stability early warning method and system, computing device and medium

Abstract

The invention discloses a power system voltage stability early warning method and system, a computing device and a medium, relates to the technical field of power grid safety risk control, and solves the problem that in the prior art, voltage stability risks cannot be early warned and avoided through post analysis. According to the method, the predicted target variable at the future moment is predicted based on the target variable and the associated variable in the operation data of the power system, then the power system is simulated based on the predicted target variable, and then the voltage stability of the power system at the future moment is analyzed based on the simulated operation data. And the early warning action is executed based on the analysis result, so that the voltage stability risk of the power system can be early warned and avoided in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid security risk control, and particularly to a voltage stability early warning method, system, computing device and medium for a power system. Background Art

[0002] The energy and environmental protection crises have made new energy sources such as solar energy and wind energy increasingly attracting people's attention. New energy power generation has gradually become the trend of power development. In recent years, new energy has been connected to the power grid on a large scale year by year. However, the output of new energy power generation is easily affected by weather changes, seasonal variations, and geographical locations, etc., with intermittency and uncertainty, and the power generation power has large fluctuations. Since new energy sources such as wind power and solar energy have random and intermittent characteristics, they will have a certain impact on the power grid after being connected to the grid. New energy generating units often adopt power generation technologies different from traditional synchronous generating units, and their steady-state and transient characteristics are different from those of traditional synchronous generating units. After a high proportion of new energy is connected to the grid and relatively concentrated in a certain regional power grid, the voltage stability and transient stability of the power grid will change. The voltage safety problem is the most serious threat to the safe operation of the power grid brought by the high proportion of new energy connected to the grid.

[0003] Traditional voltage stability analysis methods are mainly ex post analysis, and cannot early warn and avoid voltage stability risks. In view of this, a voltage stability early warning method, system, computing device and medium for a power system are needed. Summary of the Invention

[0004] Aiming at the problem that ex post analysis in the prior art cannot early warn and avoid voltage stability risks, the present invention provides a voltage stability early warning method, system, computing device and medium for a power system, which can early warn and avoid voltage stability risks.

[0005] In a first aspect, an embodiment of the present application provides a voltage stability early warning method for a power system, including:

[0006] Obtain the operation data of the power system, and construct a data set containing operation variables based on the operation data; screen out associated variables from the non-target variables based on correlation analysis; input the associated variables and the target variable into a preset long short-term memory (LSTM) model to obtain the predicted target variable output by the LSTM model; perform simulation of the power system based on the predicted target variable to obtain simulation operation data; analyze the voltage stability of the power system at a future moment based on the simulation operation data to obtain the voltage stability risk level of the power system; and perform an early warning action corresponding to the voltage stability risk level based on the voltage stability risk level.

[0007] Among them, the data set includes a target variable and non-target variables. The target variable is used to simulate the operation of the power system; the correlation variable has a correlation value with the target variable greater than a preset correlation threshold, and the correlation value is used to indicate the correlation between two variables; the predicted target variable is the target variable of the power system at a future moment predicted by the LSTM model.

[0008] Preferably, the target variables include active power, reactive power, and the output of new energy units; the correlation variables include meteorological factors and time factors.

[0009] Preferably, the screening of the correlation variables from the non-target variables based on correlation analysis includes: calculating the mutual information between the non-target variables and the target variable; determining the correlation variables based on the mutual information and the correlation threshold; the inputting of the correlation variables and the target variable into a preset LSTM model includes: by using the principal component analysis method, projecting the data matrix composed of the correlation variables and the target variable into a low-dimensional space to obtain a reduced-dimensional matrix; inputting the reduced-dimensional matrix into the LSTM model.

[0010] Preferably, the simulation operation data includes bus voltage and phase angle; the analysis of the voltage stability of the power system at a future moment based on the simulation operation data to obtain the voltage stability risk level of the power system includes: calculating the minimum singular value of the Jacobian matrix based on the bus voltage, the phase angle, the active power and the reactive power in the predicted target variable; determining the voltage stability risk level based on the magnitude relationship between the minimum singular value and the warning threshold.

[0011] Preferably, the warning threshold includes a first threshold, a second threshold, and a third threshold from small to large; when the minimum singular value is less than the first threshold, the voltage stability risk level is at the normal level, and the corresponding warning action is to maintain the current state of the power system; when the minimum singular value is greater than or equal to the first threshold and less than the second threshold, the voltage stability risk level is at the attention level; when the minimum singular value is greater than or equal to the second threshold and less than the third threshold, the voltage stability risk level is at the warning level; when the minimum singular value is greater than or equal to the third threshold, the voltage stability risk level is at the severe level.

[0012] Preferably, the execution of the warning action corresponding to the voltage stability risk level based on the voltage stability risk level includes: when the voltage stability risk level is at the normal level, maintaining the current state of the power system; when the voltage stability risk level is at the attention level, adjusting the load distribution of the power system; when the voltage stability risk level is at the warning level, starting the standby power supply; when the voltage stability risk level is at the severe level, cutting off the load of the power system.

[0013] Preferably, based on the voltage stability risk level, warning actions corresponding to the voltage stability risk level are executed, including: when the voltage stability risk level is at the normal level, outputting a first warning message suggesting maintaining the current state of the power system; when the voltage stability risk level is at the concerned level, outputting a second warning message suggesting adjusting the load distribution of the power system; when the voltage stability risk level is at the warning level, outputting a third warning message suggesting starting the standby power supply; when the voltage stability risk level is at the severe level, outputting a fourth warning message suggesting cutting off the load of the power system.

[0014] Preferably, before screening the associated variables from the non-target variables based on the correlation analysis, the method further includes: supplementing the data set based on the interpolation method; detecting outliers in the data set based on the Z-Score method and deleting the outliers from the data set; normalizing the data set after data supplementation and outlier removal.

[0015] In a second aspect, an embodiment of the present application provides a power system voltage stability warning system, which is applied to the method in the first aspect. The system includes:

[0016] An acquisition module, configured to acquire the operation data of the power system, and construct a data set including operation variables based on the operation data. The data set includes a target variable and non-target variables, and the target variable is used to simulate the operation of the power system;

[0017] A first analysis module, configured to screen associated variables from the non-target variables based on correlation analysis. The correlation value between the associated variable and the target variable is greater than a preset correlation threshold, and the correlation value is used to indicate the correlation between two variables;

[0018] A prediction module, configured to input the associated variable and the target variable into a preset long short-term memory (LSTM) model, and obtain a predicted target variable output by the LSTM model. The predicted target variable is the target variable of the power system predicted by the LSTM model at a future moment;

[0019] A simulation module, configured to perform simulation of the power system based on the predicted target variable to obtain simulation operation data;

[0020] A second analysis module, configured to analyze the voltage stability of the power system at a future moment based on the simulation operation data to obtain the voltage stability risk level of the power system;

[0021] A warning module, configured to execute warning actions corresponding to the voltage stability risk level based on the voltage stability risk level.

[0022] In a third aspect, an embodiment of the present application provides a computing device, including: a memory for storing a program; and a processor for loading the program to execute the method described in the first aspect.

[0023] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute the method described in the first aspect.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows: By predicting the predicted target variable at a future moment based on the target variable and associated variables in the power system operation data, then performing simulation of the power system based on the predicted target variable, and then performing voltage stability analysis on the simulated circuit system, it is equivalent to performing voltage stability analysis on the power system at a future moment in advance; and then performing corresponding warning actions based on the analysis results, which can warn and avoid the voltage stability risks of the power system in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0026] Figure 1 It is a schematic flowchart of a method for warning voltage stability of a power system provided by an embodiment of the present application;

[0027] Figure 2 It is a schematic structural diagram of a system for warning voltage stability of a power system provided by an embodiment of the present application;

[0028] Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0030] It should be understood that when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0031] It should also be understood that the terms used in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0032] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0033] To solve the problem that in the existing voltage stability analysis technology of power systems, voltage stability risks cannot be pre-warned and avoided due to ex post analysis, the present invention provides a power system voltage stability early warning method, system, computing device, and medium, which can pre-warn and avoid voltage stability risks.

[0034] Please refer to Figure 1 , Figure 1 which is a power system voltage stability early warning method provided by an embodiment of the present application. This method can be applied to a computing device; the method specifically includes the following steps:

[0035] Step 101, the computing device obtains the operation data of the power system and constructs a data set containing operation variables based on this operation data.

[0036] Among them, the computing device can be communicatively connected to the detection devices of each node in the power system in a wired or wireless manner; the detection device detects the corresponding node to obtain the operation data of the node and sends the operation data to the computing device; after the computing device receives the operation data of the corresponding node from each detection device, it can summarize and integrate these operation data to construct a data set containing operation variables.

[0037] Among them, the operation variables in this data set may include: active power P i (t), reactive power Q i (t), output of new energy units P RES,i (t), meteorological factor W n,i (t), and time factor S m,i(t). Among them, i represents the node number, t represents time, n represents the types of meteorological factors, and m represents the types of time factors. Exemplarily, the time factors may include the equipment commissioning time, lighting time, residential electricity consumption time, etc.; the meteorological factors may include light intensity, wind force level, temperature and humidity, etc.

[0038] Specifically, the output of the new energy unit refers to the electric energy output by the new energy unit per unit time.

[0039] Among them, the data set includes target variables and non-target variables. The target variables are used to simulate the operation of the power system. It can be understood that other variables in the data set except the target variables can be regarded as non-target variables.

[0040] Preferably, the target variables include active power, reactive power, and the output of the new energy unit.

[0041] Preferably, after constructing the data set, the computing device can preprocess the data set and then execute step 102 based on the preprocessed data set. Specifically, the computing device can supplement the data set based on the interpolation method, then detect the outliers in the data set based on the Z-Score method, and delete the outliers from the data set; then perform data standardization on the data set after data supplementation and outlier removal.

[0042] The calculation formula of the interpolation method is:

[0043]

[0044] Among them, t i represents the time point corresponding to the interpolated data X(t i ), t i-1 represents the previous time point of the time point t i , t i+1 represents the next time point of the time point t i ; X(t i-1 ) represents the data at the time point t i-1 , X(t i+1 ) represents the data at the time point.

[0045] The calculation formula for detecting outliers by the Z-Score method is:

[0046]

[0047] Among them, μ is the mean of the data X, and σ is the standard deviation of the data X; if |Z|>3, it is determined as an outlier.

[0048] Among them, the computing device can also align the data from different nodes based on timestamps and calculate the synchronization value using linear interpolation. The formula for calculating the synchronization value can refer to the calculation formula of the above interpolation method and will not be elaborated here.

[0049] The calculation formula for data standardization is:

[0050]

[0051] Among them, X′ is the standardized data, and X min and X max are the minimum and maximum values of the data respectively.

[0052] Step 102: The computing device screens for associated variables from the non-target variables based on correlation analysis.

[0053] Among them, in step 103, the computing device needs to predict the target variables that can be used to simulate the power system through a long short-term memory (LSTM) model. In order to reduce the computational complexity required for the model to predict the target variables at future moments and improve the prediction accuracy, it is necessary to find data pairs with strong correlations among the non-target variables in the operating data to train the model.

[0054] Among them, the computing device can directly determine the associated variables in the dataset based on the pre-completed correlation analysis results, thereby determining the input of the LSTM model.

[0055] Among them, the computing device can screen for associated variables with strong correlation relationships with the target variables among the non-target variables through methods such as mutual information method, distance correlation method, canonical correlation analysis (CCA) method, or other correlation analysis methods.

[0056] Specifically, for non-target variables whose correlation values with the target variable are greater than a preset correlation threshold, the computing device can determine the non-target variable as an associated variable. It can be understood that the correlation threshold can be preset corresponding to the adopted correlation analysis method. The correlation value is specifically an index used to indicate the correlation between two variables in various correlation analysis methods.

[0057] Preferably, the associated variables include meteorological factors and time factors. Among them, there are many meteorological factors and time factors, and power systems located in different geographical locations, with different scales, and different maintenance conditions may be affected by meteorological factors and time factors to different extents. Therefore, one or more meteorological factors and time factors that have a greater impact on the power system can be selected as the associated variables.

[0058] Preferably, the computing device filters the associated variables from the non-target variables through the mutual information method, which specifically includes: calculating the mutual information between the non-target variable and the target variable; determining the associated variable based on the mutual information and the correlation threshold.

[0059] Among them, the computing device can first calculate the univariate entropy of each non-target variable, and the calculation formula is:

[0060] H(A) = -∑ a∈A p(a)logp(a)

[0061] where A is the non-target variable, a is the data corresponding to the non-target variable, H(A) is the univariate entropy of A, and p(a) is the probability distribution of A.

[0062] Then, the computing device can calculate the joint entropy of the non-target variable and the target variable under different combinations, and the calculation formula is:

[0063] H(A,B) = -∑ a∈A,b∈B p(a,b)logp(a,b)

[0064] where B is the target variable, y is the data corresponding to the target variable, and H(A,B) is the joint entropy of A and B.

[0065] Then, the computing device can calculate the mutual information between the non-target variable and the target variable, and the calculation formula is:

[0066] I(A;B) = H(A) + H(B) - H(A,B)

[0067] where I(A;B) is the mutual information between the non-target variable A and the target variable B.

[0068] Then, the computing device can normalize the mutual information to better represent the correlation. The calculation formula is:

[0069]

[0070] Among them, NMI(A; B) is the mutual information between the non-target variable A and the target variable B after normalization. Specifically, the range of the mutual information after normalization is between [0, 1], and the larger the value, the stronger the correlation. At this time, the correlation threshold can be set to 0.3. When the mutual information is higher than 0.3, the computing device can determine the corresponding non-target variable as the associated variable.

[0071] After determining the associated variable, the computing device can obtain a data matrix including the associated variable and the target variable, reduce the dimension of the data matrix through a data dimensionality reduction method, and then use the data matrix after dimensionality reduction as the input of the LSTM model. The data dimensionality reduction method can further reduce the computational amount of the model while maintaining the characteristics of the data matrix.

[0072] Among them, the computing device can use linear dimensionality reduction methods such as principal component analysis (PCA), linear discriminant analysis (LDA), or factor analysis (FA), or can also use non-linear dimensionality reduction methods such as t-distributed stochastic neighbor embedding method, Isomap method, or locally linear embedding method to reduce the dimension of the data matrix.

[0073] Preferably, the computing device can project the data matrix composed of the associated variable and the target variable into a low-dimensional space through the principal component analysis method to obtain a dimensionality reduction matrix, and then input the dimensionality reduction matrix into the LSTM model.

[0074] Among them, the computing device can first calculate the covariance matrix of the data matrix X, and the calculation formula is:

[0075]

[0076] Among them, C represents the covariance matrix, n is the number of samples in the data matrix X, is the transpose matrix of X.

[0077] Then, the computing device can decompose the eigenvalues of the covariance matrix, and the calculation formula is:

[0078] Cv = λv

[0079] Among them, λ is the eigenvalue and v is the corresponding eigenvector.

[0080] Then, the computing device can calculate the cumulative variance contribution rate, and the calculation formula of the cumulative variance contribution rate is:

[0081]

[0082] Among them, p is the number of eigenvalues of the covariance matrix, and usually η = 95% is taken.

[0083] Then, the computing device can select the eigenvectors corresponding to the top k eigenvalues in the order of contribution rate from high to low for data projection. The calculation formula is:

[0084] X 降维 = XV k

[0085] Among them, X 降维 represents the data matrix after dimensionality reduction, and V k is the matrix composed of the top k eigenvectors.

[0086] Step 103: The computing device inputs the correlation variable and the target variable into a preset long short-term memory (LSTM) model to obtain the predicted target variable output by the LSTM model.

[0087] Among them, after obtaining the dimensionality reduction matrix in Step 102, the computing device can input the dimensionality reduction data matrix into the LSTM model.

[0088] Among them, the LSTM model includes an input layer, a hidden layer, and an output layer. The input layer is used to receive the multi-dimensional dimensionality reduction matrix; the hidden layer includes multiple LSTM cells, which can capture the dynamic characteristics of the time series; the output layer is used to predict the target variable at a future moment. The predicted target variable is the target variable of the power system predicted by the LSTM model at a future moment.

[0089] Preferably, the predicted target variable output by the LSTM model includes the active power P i (t + Δt), reactive power Q i (t + Δt), and new energy output P RES,i (t + Δt).

[0090] Among them, the LSTM model is a pre-trained LSTM model set in the computing device. The LSTM model can be trained based on the historical operation data of one or more power systems, define the loss function of the mean square error, and update the model parameters using the backpropagation algorithm and an optimizer (such as Adam).

[0091] Specifically, the historical data can include the normal operation data of the power system, as well as the operation data before and after the occurrence of voltage stability problems; the historical data can also include annotations for marking the occurrence time and severity of voltage stability events.

[0092] Specifically, the loss function of the mean square error is as follows:

[0093]

[0094] where L is the loss value, N is the number of samples, and t i represents the time point, represents, represents.

[0095] Step 104: The computing device performs simulation of the power system based on the predicted target variable to obtain simulation operation data.

[0096] Specifically, the computing device may input the predicted target variable into simulation software to simulate the operation of the power system and obtain simulation operation data.

[0097] Specifically, the simulation operation data may include bus voltage V i (t + Δt), line current I ij (t + Δt), and frequency f(t) + Δt.

[0098] Step 105: The computing device analyzes the voltage stability of the power system at a future time based on the simulation operation data to obtain the voltage stability risk level of the power system.

[0099] Specifically, the computing device may calculate an index S reflecting voltage stability from the simulation operation data.

[0100] Preferably, the simulation operation data includes bus voltage and phase angle; the computing device may calculate the minimum singular value of the Jacobian matrix based on the bus voltage, the phase angle, the active power and the reactive power in the predicted target variable; and determine the voltage stability risk level based on the magnitude relationship between the minimum singular value and the warning threshold.

[0101] The calculation formula of the Jacobian matrix J is as follows:

[0102]

[0103] S = σ min = min(singular values of J)

[0104] where J is the Jacobian matrix of the simulated power system, which describes the dynamic characteristics of the simulated power system, and σ min is the minimum singular value, representing the minimum eigenvalue of the Jacobian matrix, serving as the stability index S. P is the active power in the predicted target variable, Q is the reactive power in the predicted target variable, V is the bus voltage in the simulation operation data, and δ is the phase angle in the simulation operation data.

[0105] The Jacobian matrix is a first-order derivative matrix with respect to active power, reactive power, voltage amplitude, and phase angle, reflecting the linearized characteristics of the simulated power system near the equilibrium point; this minimum singular value reflects the sensitivity of the system to disturbances. When the system approaches the voltage collapse point, this minimum singular value approaches zero. Therefore, the sensitivity of the system to disturbances can be quantified by the minimum singular value of the Jacobian matrix, and this minimum singular value is used as an index to reflect the voltage stability risk.

[0106] Preferably, the warning threshold includes a first threshold, a second threshold, and a third threshold from small to large; when the minimum singular value is less than the first threshold, the voltage stability risk level is the normal level, and the corresponding warning action is to maintain the current state of the power system; when the minimum singular value is greater than or equal to the first threshold and less than the second threshold, the voltage stability risk level is the concerned level; when the minimum singular value is greater than or equal to the second threshold and less than the third threshold, the voltage stability risk level is the warning level; when the minimum singular value is greater than or equal to the third threshold, the voltage stability risk level is the severe level.

[0107] Among them, the warning threshold can be set based on grid operation regulations and safety standards, or the warning threshold can be dynamically adjusted based on the actual operation conditions of the grid.

[0108] Specifically, the grid operation regulations and safety standards can include "Guide for Safety and Stability of Power Systems", "Specification for Maintaining Stability Thresholds of Power Systems", and international IEC series specifications, etc.

[0109] Specifically, the warning threshold can be adjusted based on the actual load of the grid; when the load of the power system is low, the warning threshold can be increased; when the load is high, the warning threshold can be decreased.

[0110] Step 106, the computing device performs a warning action corresponding to the voltage stability risk level based on the voltage stability risk level.

[0111] Among them, after determining the voltage stability risk level of the power system, the computing device can perform the corresponding warning action.

[0112] Optionally, in the case where the power system can be directly controlled, the computing device can directly perform an operation that can reduce the voltage stability risk.

[0113] Among them, when the voltage stability risk level is at the normal level, the computing device can maintain the current state of the power system; when the voltage stability risk level is at the concerned level, the computing device can adjust the load distribution of the power system; when the voltage stability risk level is at the warning level, the computing device can start the standby power supply; when the voltage stability risk level is at the severe level, the computing device can cut off the load of the power system.

[0114] Exemplarily, the computing device can adjust the load distribution of the power system based on a preset control strategy, such as adjusting based on the goal of load balancing, or based on the goal of reducing the load of weak nodes, or based on the goal of balancing reactive power load in hierarchical and zonal areas. The adjustment formula is:

[0115] P' = P - ΔP, Q' = Q + ΔQ

[0116] Where P and Q are the active power and reactive power respectively, ΔP and ΔQ are the adjustment amounts of active power and reactive power respectively, and P’ and Q’ are the adjusted active power and reactive power respectively.

[0117] Exemplarily, the computing device can determine the power to be supplemented based on the power consumption time or the importance level of the power supply object, so as to determine the standby power supply to be enabled. The power formula is:

[0118]

[0119] Where, is the power of each standby power supply, P 备用 is the overall power of the enabled standby power supply.

[0120] Exemplarily, the computing device can determine the nodes capable of cutting off the load, the power amplitude that can be cut off, and the power amplitude required to maintain voltage stability based on the power consumption time or the importance level of the power supply object, so as to determine the target nodes and the load amplitude to be cut off for each target node. The cutting formula is:

[0121]

[0122] Where, is the cutting power of each target node, P 切除 is the total power to be cut off.

[0123] Optionally, when the power system cannot be directly controlled, the computing device can output an alarm message to an output device that can be perceived by the maintenance personnel or management personnel of the power system, as well as operation suggestions for reducing the voltage stability risk.

[0124] Exemplarily, when the voltage stability risk level is at the normal level, the computing device may output a first warning message suggesting maintaining the current state of the power system; when the voltage stability risk level is at the concerned level, the computing device may output a second warning message suggesting adjusting the load distribution of the power system; when the voltage stability risk level is at the warning level, the computing device may output a third warning message suggesting starting the backup power supply; when the voltage stability risk level is at the severe level, the computing device may output a fourth warning message suggesting cutting off the load of the power system.

[0125] The present invention predicts a predicted target variable at a future moment based on a target variable and associated variables in the power system operation data, then performs a simulation of the power system based on the predicted target variable, and then conducts a voltage stability analysis on the simulated circuit system, which is equivalent to performing a voltage stability analysis on the power system at a future moment in advance; and then executes corresponding warning actions based on the analysis results, capable of early warning and avoiding the voltage stability risk of the power system.

[0126] The present invention is applicable to power systems with a high proportion of new energy access, especially regional power grids with a relatively high proportion of new energy and complex operation modes. Such power grids have greater randomness and volatility and more non-linear parameters; the present invention performs stability prediction through an LSTM network model, can better adapt to the characteristics of such power grids, learn their non-linear characteristics, thus more accurately predict the voltage stability risk, improve its ability to cope with voltage stability risks, and ensure the safe and stable operation of the power system.

[0127] The method provided in the embodiments of the present application has been described above. Next, the related devices provided by the present application will be described.

[0128] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of a power system voltage stability warning system provided by the embodiments of the present application. As Figure 2 shown, the system 200 includes:

[0129] An acquisition module 201, configured to acquire the operation data of the power system, and construct a data set including operation variables based on the operation data. The data set includes a target variable and non-target variables, and the target variable is used to simulate the operation of the power system;

[0130] A first analysis module 202, configured to screen associated variables from the non-target variables based on correlation analysis. The correlation value between the associated variable and the target variable is greater than a preset correlation threshold, and the correlation value is used to indicate the correlation between two variables;

[0131] A prediction module 203, configured to input the correlation variable and the target variable into a preset long short-term memory (LSTM) model, and obtain a predicted target variable output by the LSTM model, where the predicted target variable is the target variable of the power system predicted by the LSTM model at a future moment;

[0132] A simulation module 204, configured to perform simulation on the power system based on the predicted target variable to obtain simulation operation data;

[0133] A second analysis module 205, configured to analyze the voltage stability of the power system at a future moment based on the simulation operation data to obtain a voltage stability risk level of the power system;

[0134] An early warning module 206, configured to perform an early warning action corresponding to the voltage stability risk level based on the voltage stability risk level.

[0135] Preferably, the target variable includes active power, reactive power, and the output of new energy units; the correlation variable includes meteorological factors and time factors.

[0136] Preferably, the first analysis module 202 is specifically configured to calculate the mutual information between the non-target variable and the target variable; determine the correlation variable based on the mutual information and the correlation threshold; the prediction module 203 is specifically configured to project a data matrix formed by the correlation variable and the target variable into a low-dimensional space through a principal component analysis method to obtain a dimensionality reduction matrix; and input the dimensionality reduction matrix into the LSTM model.

[0137] Preferably, the simulation operation data includes bus voltage and phase angle; the second analysis module 205 is specifically configured to calculate the minimum singular value of the Jacobian matrix based on the bus voltage, the phase angle, the active power, and the reactive power in the predicted target variable; and determine the voltage stability risk level based on the magnitude relationship between the minimum singular value and the early warning threshold.

[0138] Preferably, the early warning threshold includes a first threshold, a second threshold, and a third threshold from small to large; when the minimum singular value is less than the first threshold, the voltage stability risk level is at a normal level, and the corresponding early warning action is to maintain the current state of the power system; when the minimum singular value is greater than or equal to the first threshold and less than the second threshold, the voltage stability risk level is at a concerned level; when the minimum singular value is greater than or equal to the second threshold and less than the third threshold, the voltage stability risk level is at a warning level; when the minimum singular value is greater than or equal to the third threshold, the voltage stability risk level is at a severe level.

[0139] Preferably, the early warning module 206 is specifically configured to maintain the current state of the power system when the voltage stability risk level is at the normal level; adjust the load distribution of the power system when the voltage stability risk level is at the concerned level; start the standby power supply when the voltage stability risk level is at the warning level; and cut off the load of the power system when the voltage stability risk level is at the severe level.

[0140] Preferably, the early warning module 206 is specifically configured to output a first early warning message suggesting to maintain the current state of the power system when the voltage stability risk level is at the normal level; output a second early warning message suggesting to adjust the load distribution of the power system when the voltage stability risk level is at the concerned level; output a third early warning message suggesting to start the standby power supply when the voltage stability risk level is at the warning level; and output a fourth early warning message suggesting to cut off the load of the power system when the voltage stability risk level is at the severe level.

[0141] Preferably, the system 200 further includes: a preprocessing module 207, which is configured to supplement the data set based on the interpolation method; detect outliers in the data set based on the Z-Score method and delete the outliers from the data set; and perform data standardization on the data set after data supplementation and outlier removal.

[0142] The power system voltage stability early warning system provided by the embodiments of the present application can be understood by referring to the corresponding content in the foregoing method embodiment part, and will not be repeated here.

[0143] Please refer to Figure 3 , such as Figure 3 shown, Figure 3 is a possible logical structure diagram of a computing device provided by the embodiments of the present application. The computing device 300 includes: a processor 301, a communication interface 302, a memory 303, and a bus 304. The processor 301, the communication interface 302, and the memory 303 are interconnected through the bus 304. In the embodiments of the present application, the processor 301 is used to control and manage the actions of the computing device 300. For example, the processor 301 is used to execute Figure 1 the steps in the embodiments and / or other processes for the technologies described herein. The communication interface 302 is used to support the communication of the computing device 300. The memory 303 is used to store the program code and data of the computing device 300.

[0144] Among them, the processor 301 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on. The bus 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 3 only a thick line is used to represent it in Figure 3 , but it does not mean that there is only one bus or one type of bus.

[0145] In another embodiment of the present application, a computer-readable storage medium is further provided. The computer-readable storage medium includes instructions that, when run on a computer, cause the computer to execute the method described in the above-mentioned Embodiment 1.

[0146] Those of ordinary skill in the art can realize that the units of the examples described in connection with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components of the examples have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0147] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0148] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0149] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0150] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0151] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The foregoing storage medium includes: USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs and other various media that can store program codes.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the claims and the description of the present invention.

Claims

1. A method for early warning of voltage stability in a power system, characterized in that, Including: Obtain the operation data of the power system, and construct a data set including operation variables based on the operation data. The operation variables include target variables and non-target variables, and the target variables are used to simulate the operation of the power system. Screen out associated variables from the non-target variables based on correlation analysis. The correlation value between the associated variables and the target variables is greater than a preset correlation threshold, and the correlation value is used to indicate the correlation between two variables. Input the associated variables and the target variables into a preset long short-term memory network (LSTM) model to obtain the predicted target variables output by the LSTM model. The predicted target variables are the target variables of the power system predicted by the LSTM model at future times. Perform simulation of the power system based on the predicted target variables to obtain simulation operation data. Analyze the voltage stability of the power system at the future time based on the simulation operation data to obtain the voltage stability risk level of the power system. Execute a warning action corresponding to the voltage stability risk level based on the voltage stability risk level.

2. The method according to claim 1, wherein The target variables include active power, reactive power, and the output of new energy units; the associated variables include meteorological factors and time factors.

3. The method according to claim 1 or 2, characterized in that, The screening of associated variables from the non-target variables based on correlation analysis includes: Calculate the mutual information between the non-target variables and the target variables. Determine the associated variables based on the mutual information and the correlation threshold. The inputting of the associated variables and the target variables into a preset long short-term memory network (LSTM) model includes: By using the principal component analysis method, project the data matrix composed of the associated variables and the target variables into a low-dimensional space to obtain a dimensionality-reduced matrix. Input the dimensionality-reduced matrix into the LSTM model.

4. The method according to claim 1 or 2, characterized in that, The simulation operation data includes bus voltage and phase angle; the analysis of the voltage stability of the power system at the future time based on the simulation operation data to obtain the voltage stability risk level of the power system includes: Calculate the minimum singular value of the Jacobian matrix based on the bus voltage, the phase angle, the active power, and the reactive power in the predicted target variables. Determine the voltage stability risk level based on the magnitude relationship between the minimum singular value and a preset warning threshold.

5. The method according to claim 1 or 2, characterized in that, The execution of a warning action corresponding to the voltage stability risk level based on the voltage stability risk level includes: Under the condition that the voltage stability risk level is at the normal level, maintain the current state of the power system. Under the condition that the voltage stability risk level is at the concerned level, adjust the load distribution of the power system. Under the condition that the voltage stability risk level is at the warning level, start the standby power supply. Under the condition that the voltage stability risk level is at the severe level, cut off the load of the power system.

6. The method according to claim 1 or 2, characterized in that, The execution of a warning action corresponding to the voltage stability risk level based on the voltage stability risk level includes: When the voltage stability risk level is at the normal level, output the first warning information suggesting to maintain the current state of the power system. When the voltage stability risk level is at the concerned level, output the second warning information suggesting to adjust the load distribution of the power system. When the voltage stability risk level is at the warning level, output the third warning information suggesting to start the backup power supply. When the voltage stability risk level is at the severe level, output the fourth warning information suggesting to cut off the load of the power system.

7. The method according to claim 1 or 2, characterized in that Before screening the associated variables from the non-target variables based on the correlation analysis, the method further includes: Supplementary data to the dataset based on the interpolation method. Detect outliers in the dataset based on the Z-Score method and delete the outliers from the dataset. Perform data standardization processing on the dataset after completing data supplementation and removing outliers.

8. A power system voltage stability early warning system, characterized in that, Applied to any of the methods described in claims 1-7, the system includes: An acquisition module, configured to acquire the operation data of the power system, and construct a dataset including operation variables based on the operation data. The dataset includes a target variable and non-target variables, and the target variable is used to simulate the operation of the power system. A first analysis module, configured to screen associated variables from the non-target variables based on correlation analysis. The correlation value between the associated variable and the target variable is greater than a preset correlation threshold, and the correlation value is used to indicate the correlation between two variables. A prediction module, configured to input the associated variable and the target variable into a preset long short-term memory (LSTM) model, and obtain the predicted target variable output by the LSTM model. The predicted target variable is the target variable of the power system predicted by the LSTM model at a future time. A simulation module, configured to perform simulation of the power system based on the predicted target variable to obtain simulation operation data. A second analysis module, configured to analyze the voltage stability of the power system at the future time based on the simulation operation data, and obtain the voltage stability risk level of the power system. A warning module, configured to perform a warning action corresponding to the voltage stability risk level based on the voltage stability risk level.

9. A computing device, characterized in that, Includes: A memory, configured to store a program. A processor, configured to load the program to execute any of the methods described in claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein when the program runs, it controls the device where the computer-readable storage medium is located to execute any of the methods described in claims 1-7.

Images