A wide-area monitoring method for subsynchronous oscillation of new energy power grid based on Internet of Things

Abstract

The present invention relates to the technical field of subsynchronous oscillation monitoring, and specifically to a wide-area monitoring method for subsynchronous oscillation of a new energy power grid based on the Internet of Things, comprising: setting a wide-area monitoring substation at each node in the target new energy power grid, and a plurality of the wide-area monitoring substations communicate data with a wide-area monitoring main station, and the wide-area monitoring substations collect phasor data of the nodes in the target new energy power grid through a PMU to obtain the phasor data. By setting a wide-area monitoring substation at each node and communicating data with the main station, the present invention can obtain the phasor data of each node in real time, which helps to fully grasp the dynamic behavior of the power grid, and by mapping the phasor data of the node with the network topology diagram in real time, it can accurately reflect the flow direction and state of the phasor in the power grid, and enhance the real-time monitoring capability of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of subsynchronous oscillation monitoring, and in particular to a wide-area monitoring method for subsynchronous oscillation of a new energy power grid based on the Internet of Things. Background Art

[0002] PMU is a device used to measure the phasor data of voltage and current in the power grid in real time. It provides accurate phase and amplitude information through synchronous measurement technology. PMU is mainly used to monitor the dynamic behavior of the power grid, such as subsynchronous oscillation and system stability. The Internet of Things can transmit the data generated by these PMUs to the central system through the network to achieve real-time data aggregation and analysis. The Internet of Things technology can make PMU data more widely accessible and analyzed, improving data availability and monitoring scope.

[0003] Many traditional methods rely on offline data analysis, that is, the data needs to be stored and processed after collection, which leads to the lack of real-time monitoring capabilities and the inability to quickly respond to subsynchronous oscillations in the power grid. Due to the limitations of traditional data collection and transmission systems, the collected data may have time delays, making the real-time status monitoring not timely enough. In addition, many traditional methods need to rely on manual data analysis and anomaly identification, which not only increases the possibility of human error, but also makes the processing process inefficient. Moreover, in traditional methods, the feature extraction process may be more complicated and require manual configuration and adjustment, which makes identifying anomalies in the power grid status cumbersome and error-prone. Summary of the invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and to propose a wide-area monitoring method for subsynchronous oscillation of a new energy power grid based on the Internet of Things.

[0005] The technical solution of the present invention is a method for monitoring subsynchronous oscillation of a new energy power grid in a wide area based on the Internet of Things, comprising:

[0006] A wide area monitoring substation is set at each node in the target new energy grid, and a plurality of the wide area monitoring substations communicate data with a wide area monitoring master station, and the wide area monitoring substations collect phasor data of the nodes in the target new energy grid through a PMU to obtain the phasor data;

[0007] Acquire a network topology diagram of the target new energy power grid, and map the phasor data of each node of the target new energy power grid with the network topology diagram in real time to obtain a phasor flow network topology diagram of the target new energy power grid;

[0008] The wide area monitoring substation compresses the phasor data to obtain standard phasor data of nodes in the target new energy grid, and uploads the standard phasor data of nodes in the target new energy grid to the wide area monitoring main station;

[0009] The wide area monitoring master station calculates a unified oscillation frequency of nodes in the target new energy grid based on standard phasor data of the nodes in the target new energy grid, and compares the unified oscillation frequency with a preset frequency threshold;

[0010] If the error between the two exceeds a preset frequency range, parameter identification is performed on the standard phasor data of the node in the target new energy grid to obtain identification parameters of the node in the target new energy grid, and the identification parameters are compared with a preset parameter threshold;

[0011] If the parameter error between the two exceeds the preset parameter range, feature extraction is performed on the standard phasor data to obtain a feature set corresponding to the standard phasor data, and the feature set is input into a trained state classification model, and the state of the node in the target new energy power grid is output through the trained state classification model.

[0012] Preferably, the phasor data of each node of the target new energy grid is mapped in real time with the network topology diagram to obtain the phasor flow network topology diagram of the target new energy grid, including:

[0013] Using a power system modeling tool to graphically model the network topology of the target new energy power grid to obtain a network topology model;

[0014] Creating a node mapping table to map the plurality of wide area monitoring substations to corresponding nodes in the network topology model;

[0015] Based on the node mapping table, mapping the phasor data of each node of the target new energy power grid with the corresponding node in the network topology model;

[0016] The voltage amplitude, phase angle and current flow direction of each node in the network topology model are represented based on color coding, size changes of voltage amplitude symbols and arrows indicating current flow direction to obtain a phasor flow network topology diagram of the target new energy power grid.

[0017] Preferably, the wide area monitoring substation compresses the phasor data to obtain standard phasor data of nodes in the target new energy grid, including:

[0018] Inputting the phasor data into a trained generation model, and compressing the phasor data through the trained generation model to obtain standard phasor data of the target new energy power grid;

[0019] The generative model adopts a neural network, and the generative model includes an input layer, a hidden layer and an output layer. The input layer is used to input the phasor data, the hidden layer is used to extract hidden features of the phasor data, and the output layer is used to output reconstructed data corresponding to the phasor data based on the hidden features. The generative model adopts a logistic activation function. If the reconstructed data output by the generative model is different from the phasor data, the generative model is optimized by a loss function.

[0020] Preferably, the loss function is as follows:

[0021] ;

[0022] in, represents the loss function, Indicates the total length of the data window, The data window represents the phasor data of the node in the target new energy grid The input signal, Represents the data window in the reconstructed data The output signal, represents the weighted regularization term, represents the sparse regularization term, represents the weight coefficient of the sparse regularization term, Represents the weight coefficient of the weighted regularization term.

[0023] Preferably, the unified oscillation frequency calculation formula of the nodes in the target new energy grid is as follows:

[0024] ;

[0025] in, Indicates the first The uniform oscillation frequency of the nodes, Represents the total number of nodes in the target new energy grid, Indicates the first The modal voltage amplitude measured at each node, Indicates Nodes and The line current between the nodes, represents the weighted current weight, Indicates the set threshold value. Indicates the first The oscillation frequency measured by each node.

[0026] Preferably, performing parameter identification on the standard phasor data of the nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid includes:

[0027] The standard phasor data of the nodes in the target new energy grid are subjected to continuous wavelet transform to obtain the time series data corresponding to the standard phasor data of the nodes in the target new energy grid. The continuous wavelet transform formula is as follows:

[0028] ;

[0029] in, represents the continuous wavelet transform formula, Represents standard phasor data, represents the scale factor, represents the translation factor, represents the conjugate wavelet basis;

[0030] The time series data corresponding to the standard phasor data of the nodes in the target new energy grid is divided into a plurality of continuous frequency intervals, and the frequency interval expression is as follows:

[0031] ;

[0032] in, represents the frequency interval expression, Indicates the time series data corresponding to The frequency of

[0033] The discrete wavelet transform quantities of the plurality of continuous frequency intervals are calculated, and the discrete wavelet transform quantity calculation formula is as follows:

[0034] ;

[0035] in, represents the discrete wavelet transform quantity, represents the change in frequency, Indicates the change in scale factor;

[0036] The time series data is reconstructed based on the discrete wavelet transform amount, and the reconstruction formula of the time series data is as follows:

[0037] ;

[0038] in, Represents the time series data IMT components, express discrete value of .

[0039] Preferably, performing parameter identification on the standard phasor data of the nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid also includes:

[0040] The standard phasor data is modeled based on the reconstructed time series data to obtain a phasor basic model. The phasor basic model expression is as follows:

[0041] ;

[0042] in, Represents the basic phasor model, represents the amplitude, represents the attenuation factor, represents the oscillation frequency, represents the damping ratio, represents the phase angle, represents the undamped frequency;

[0043] The identification parameters of the nodes in the target new energy grid are solved based on the phasor basic model to obtain the damping ratio and attenuation factor of the nodes in the target new energy grid.

[0044] Preferably, feature extraction is performed on the standard phasor data to obtain a feature set corresponding to the standard phasor data, including:

[0045] The amplitude ratio of the standard phasor data is feature extracted by a first formula to obtain a first feature corresponding to the standard phasor data. The first formula is as follows:

[0046] ;

[0047] in, represents the amplitude ratio of the standard phasor data, Indicates The amplitude ratio of the frequency components, It represents the ratio of the amplitude of the maximum frequency component to the amplitude of the fundamental component. Indicates the number of frequency components in standard phasor data;

[0048] The envelope relative amplitude of the standard phasor data is feature extracted by a second formula to obtain a second feature corresponding to the standard phasor data. The second formula is as follows:

[0049] ;

[0050] middle, Represents the relative amplitude of the envelope of the standard phasor data, Represents the average value of each point in the upper envelope maximum point sequence, represents the average value of each point in the sequence of upper envelope minimum points, Represents the average value of each point on the upper envelope;

[0051] The envelope extreme value points of the standard phasor data are feature extracted by a third formula to obtain a third feature corresponding to the standard phasor data. The third formula is as follows:

[0052] ;

[0053] in, Indicates the number of envelope extreme points of standard phasor data, Indicates the length of the upper envelope maximum point sequence, Indicates the length of the sequence of minimum points of the upper envelope;

[0054] The first feature, the second feature and the third feature are combined to obtain a feature set corresponding to the standard phasor data.

[0055] Preferably, the state classification model adopts an improved support vector machine model, and the state classification model improves the support vector machine model through a grey wolf optimization algorithm.

[0056] Preferably, the support vector machine model is improved by using a grey wolf optimization algorithm, including:

[0057] The initialization parameters of the gray wolf optimization algorithm are set to initialize the position of the gray wolf population. The gray wolf population position initialization formula is as follows:

[0058] ;

[0059] in, and Indicates The iteration and The position of the gray wolf individual at the iteration, Represents the parameters of the preset control nonlinear mapping;

[0060] Calculate the fitness value of the gray wolf individual and calculate the elite individual position. The elite individual position calculation formula is as follows:

[0061] ;

[0062] in, Indicates The iteration The reverse solution of the gray wolf individual, Indicates The iteration The elite solution for the gray wolf individual, Representation interval The random number inside Indicates The maximum position of the gray wolf individual at the iteration, Indicates The minimum position of the individual gray wolf at the iteration;

[0063] Marking a preset number of the gray wolf individuals with the best fitness values ​​as leader wolves, and updating the positions of the remaining gray wolf individuals in the gray wolf population;

[0064] Calculate the updated individual fitness values ​​of the gray wolves and sort them, obtain the elite gray wolves, and update the positions of the elite gray wolves;

[0065] Repeat the above steps until the optimal solution is found or the maximum number of iterations of the gray wolf optimization algorithm is reached, so as to obtain the optimal penalty factor and kernel function parameters of the support vector machine model, and improve the support vector machine model based on the optimal penalty factor and kernel function parameters.

[0066] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0067] 1. The present invention can obtain the phasor data of each node in real time by setting up a wide-area monitoring substation at each node and communicating data with the main station, which helps to fully grasp the dynamic behavior of the power grid. By mapping the phasor data of the node with the network topology diagram in real time, the flow direction and state of the phasor in the power grid can be accurately reflected, and the real-time monitoring capability of the power grid can be enhanced. The phasor data is compressed and standardized by the wide-area monitoring substation, which reduces the amount of data, improves the efficiency of data uploading, and ensures the accuracy of the data. In addition, the wide-area monitoring main station calculates a unified oscillation frequency based on standard phasor data, which helps to monitor subsynchronous oscillation problems in the power grid and detect abnormalities in time.

[0068] 2. The present invention can quickly detect whether there is a subsynchronous oscillation risk in the power grid by comparing the calculated unified oscillation frequency with the preset frequency threshold. If the frequency error exceeds the range, the power grid state can be further analyzed through parameter identification and feature extraction, and specific oscillation characteristics can be identified, so that countermeasures can be taken in time. Moreover, by inputting the feature set into a trained state classification model, the power grid state can be automatically classified to provide intelligent state assessment. Moreover, accurate state classification can help operators quickly understand the actual situation of the power grid, make timely adjustments and maintenance decisions, and improve the safety and reliability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A schematic flow chart of the steps of an overall method in an embodiment of the present invention. DETAILED DESCRIPTION

[0070] Embodiment 1, as Figure 1 As shown, the present invention proposes a method for monitoring subsynchronous oscillation of a new energy power grid based on the Internet of Things, comprising:

[0071] S1. A wide-area monitoring substation is set at each node in the target new energy grid, and multiple wide-area monitoring substations communicate data with the wide-area monitoring master station. The wide-area monitoring substation collects phasor data of the nodes in the target new energy grid through the PMU to obtain phasor data;

[0072] S2. Obtain a network topology diagram of the target new energy power grid, and map the phasor data of each node of the target new energy power grid with the network topology diagram in real time to obtain a phasor flow network topology diagram of the target new energy power grid;

[0073] S3. The wide-area monitoring substation compresses the phasor data to obtain standard phasor data of the nodes in the target new energy grid, and uploads the standard phasor data of the nodes in the target new energy grid to the wide-area monitoring main station;

[0074] S4. The wide-area monitoring master station calculates the unified oscillation frequency of the nodes in the target new energy grid based on the standard phasor data of the nodes in the target new energy grid, and compares the unified oscillation frequency with a preset frequency threshold;

[0075] S5. If the error between the two exceeds a preset frequency range, parameter identification is performed on the standard phasor data of the node in the target new energy grid to obtain identification parameters of the node in the target new energy grid, and the identification parameters are compared with a preset parameter threshold;

[0076] S6. If the parameter error between the two exceeds the preset parameter range, feature extraction is performed on the standard phasor data to obtain a feature set corresponding to the standard phasor data, and the feature set is input into a trained state classification model, and the state of the node in the target new energy power grid is output through the trained state classification model.

[0077] In the present invention, a new energy grid refers to an electric power network that integrates renewable energy (such as wind energy, solar energy, etc.); subsynchronous oscillation is an oscillation phenomenon in an electric power system, and its frequency is lower than the fundamental frequency of the electric power system; a wide-area monitoring substation is a node in a power grid monitoring system, responsible for collecting data at various points in the power grid, and it usually includes sensors and communication equipment for collecting and transmitting power grid status information; a wide-area monitoring master station is the center of the entire monitoring system, receiving data from multiple substations, processing and analyzing the data, and providing status information of the entire power grid; phasor data refers to the phasor representation of the voltage and current of the power grid obtained by a phasor measurement unit (PMU), which provides the amplitude and phase information of the electrical quantity of each node in the power grid, and is helpful for analyzing the dynamic characteristics of the power grid; a network topology diagram is a diagram of each node in the power grid and the connection relationship between them, which shows the structure of the power grid and the layout of components; a phasor flow network topology diagram refers to a diagram generated by adding a mapping of phasor data to the network topology diagram, and this topology diagram combines the structure of the power grid and the flow direction of the electrical quantity to provide a more comprehensive view of the power grid status; data compression refers to reducing the storage space or transmission bandwidth of the original data through a certain algorithm.

[0078] Embodiment 2, a wide-area monitoring method for subsynchronous oscillation of a new energy power grid based on the Internet of Things proposed by the present invention, compared with embodiment 1, this embodiment also includes: mapping the phasor data of each node of the target new energy power grid with the network topology diagram in real time to obtain the phasor flow network topology diagram of the target new energy power grid, including:

[0079] A1. Use the power system modeling tool to perform graphical modeling on the network topology of the target new energy power grid to obtain a network topology model;

[0080] A2. Create a node mapping table to map multiple wide area monitoring substations to corresponding nodes in the network topology model;

[0081] A3. Based on the node mapping table, the phasor data of each node of the target new energy grid is mapped with the corresponding nodes in the network topology model;

[0082] A4. The voltage amplitude, phase angle and current flow direction of each node in the network topology model are represented based on color coding, size changes of voltage amplitude symbols and arrows indicating current flow direction to obtain the phasor flow network topology diagram of the target new energy power grid.

[0083] In an optional embodiment, the wide area monitoring substation compresses the phasor data to obtain standard phasor data of nodes in the target new energy grid, including:

[0084] B1. Input the phasor data into the trained generation model, and compress the phasor data through the trained generation model to obtain the standard phasor data of the target new energy power grid;

[0085] B2. The generative model adopts a neural network, and the generative model includes an input layer, a hidden layer and an output layer. The input layer is used to input phasor data, the hidden layer is used to extract hidden features of the phasor data, and the output layer is used to output the reconstructed data corresponding to the phasor data based on the hidden features. The generative model adopts a logistic activation function. If the reconstructed data output by the generative model is different from the phasor data, the generative model is optimized through a loss function.

[0086] It should be noted that the Logistic activation function is a commonly used activation function that calculates the nonlinear transformation of the input value and the output value is between 0 and 1. It is used to introduce nonlinear characteristics so that the neural network can learn complex patterns; a neural network refers to a computational model that simulates the connection structure of neurons in the human brain and is used to process and learn data. It performs complex data processing through multiple levels of nodes (neurons).

[0087] In an optional embodiment, the loss function is as follows:

[0088] ;

[0089] in, represents the loss function, Indicates the total length of the data window, The data window represents the phasor data of the node in the target new energy grid The input signal, Represents the data window in the reconstructed data The output signal, represents the weighted regularization term, represents the sparse regularization term, represents the weight coefficient of the sparse regularization term, Represents the weight coefficient of the weighted regularization term.

[0090] In an optional embodiment, the unified oscillation frequency calculation formula of the nodes in the target new energy grid is as follows:

[0091] ;

[0092] in, Indicates the first The uniform oscillation frequency of the nodes, Represents the total number of nodes in the target new energy grid, Indicates the first The modal voltage amplitude measured at each node, Indicates Nodes and The line current between the nodes, represents the weighted current weight, Indicates the set threshold value. Indicates the first The oscillation frequency measured by each node.

[0093] It should be noted that unified oscillation frequency refers to the state in which the oscillation frequency of all nodes in the power grid (such as generators or substations) remains consistent. This frequency consistency is crucial to the stability and coordination of the power grid, ensuring that all parts of the power grid can operate synchronously and avoid oscillations and instability caused by frequency differences.

[0094] In an optional embodiment, parameter identification is performed on standard phasor data of nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid, including:

[0095] C1. Perform continuous wavelet transform on the standard phasor data of the nodes in the target new energy grid to obtain the time series data corresponding to the standard phasor data of the nodes in the target new energy grid. The continuous wavelet transform formula is as follows:

[0096] ;

[0097] in, represents the continuous wavelet transform formula, Represents standard phasor data, represents the scale factor, represents the translation factor, represents the conjugate wavelet basis;

[0098] C2. Divide the time series data corresponding to the standard phasor data of the nodes in the target new energy grid into multiple continuous frequency intervals. The frequency interval expression is as follows:

[0099] ;

[0100] in, represents the frequency interval expression, Indicates the time series data corresponding to The frequency of

[0101] C3. Calculate the discrete wavelet transform amount of multiple continuous frequency intervals. The discrete wavelet transform amount calculation formula is as follows:

[0102] ;

[0103] in, represents the discrete wavelet transform quantity, represents the change in frequency, Indicates the change in scale factor;

[0104] C4. Reconstruct the time series data based on discrete wavelet transform. The reconstruction formula of time series data is as follows:

[0105] ;

[0106] in, Represents the time series data IMT components, express discrete value of .

[0107] It should be noted that continuous wavelet transform (CWT) refers to a time-frequency analysis tool used to analyze the changes of different frequency components in time series data over time.

[0108] In an optional embodiment, performing parameter identification on standard phasor data of nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid also includes:

[0109] C5. Model the standard phasor data based on the reconstructed time series data to obtain the phasor basic model. The phasor basic model expression is as follows:

[0110] ;

[0111] in, Represents the basic phasor model, represents the amplitude, represents the attenuation factor, represents the oscillation frequency, represents the damping ratio, represents the phase angle, represents the undamped frequency;

[0112] C6. Solve the identification parameters of the nodes in the target new energy grid based on the phasor basic model to obtain the damping ratio and attenuation factor of the nodes in the target new energy grid.

[0113] In an optional embodiment, feature extraction is performed on the standard phasor data to obtain a feature set corresponding to the standard phasor data, including:

[0114] D1. The amplitude ratio of the standard phasor data is feature extracted by the first formula to obtain the first feature corresponding to the standard phasor data. The first formula is as follows:

[0115] ;

[0116] in, represents the amplitude ratio of the standard phasor data, Indicates The amplitude ratio of the frequency components, It represents the ratio of the amplitude of the maximum frequency component to the amplitude of the fundamental component. Indicates the number of frequency components in standard phasor data;

[0117] D2. The envelope relative amplitude of the standard phasor data is extracted by the second formula to obtain the second feature corresponding to the standard phasor data. The second formula is as follows:

[0118] ;

[0119] in, Represents the relative amplitude of the envelope of the standard phasor data, Represents the average value of each point in the upper envelope maximum point sequence, represents the average value of each point in the sequence of upper envelope minimum points, Represents the average value of each point on the upper envelope;

[0120] D3. The third formula is used to extract the number of envelope extreme value points of the standard phasor data to obtain the third feature corresponding to the standard phasor data. The third formula is as follows:

[0121] ;

[0122] in, Indicates the number of envelope extreme points of standard phasor data, Indicates the length of the upper envelope maximum point sequence, Indicates the length of the sequence of minimum points of the upper envelope;

[0123] D4. Combine the first feature, the second feature and the third feature to obtain a feature set corresponding to the standard phasor data.

[0124] It should be noted that the envelope is a smooth curve of the signal amplitude, which represents the envelope contour of the signal. By calculating the envelope, the amplitude changes of the signal at different time points can be analyzed. The envelope can be an upper envelope (a line connecting the maximum points of the signal) and a lower envelope (a line connecting the minimum points of the signal). The relative amplitude of the envelope refers to the amplitude of the signal envelope at a certain moment relative to the amplitude at other moments. It is used to describe the amplitude characteristics and changes of the signal envelope. The extreme points of the envelope refer to the maximum and minimum points on the envelope. The maximum point is the point with the maximum amplitude on the envelope, and the minimum point is the point with the minimum amplitude on the envelope. The number of extreme points can reflect the oscillation characteristics of the signal.

[0125] In an optional embodiment, the state classification model adopts an improved support vector machine model, and the state classification model improves the support vector machine model by using a grey wolf optimization algorithm.

[0126] It should be noted that the gray wolf optimization algorithm is a nature-inspired optimization algorithm based on the hunting behavior of gray wolves. It simulates the social behavior of gray wolves in the hunting process to perform global optimization. These wolves update their positions by tracking and capturing prey, thereby finding the optimal solution in the search space; the support vector machine is a supervised learning model, mainly used for classification and regression analysis. It maximizes the classification interval of data points by finding a hyperplane. The performance of the support vector machine is affected by model parameters, such as the penalty factor and kernel function parameters.

[0127] In an optional embodiment, the support vector machine model is improved by using a gray wolf optimization algorithm, including:

[0128] E1. Set the initialization parameters of the gray wolf optimization algorithm to initialize the position of the gray wolf population. The gray wolf population position initialization formula is as follows:

[0129] ;

[0130] in, and Indicates The iteration and The position of the gray wolf individual at the iteration, Represents the parameters of the preset control nonlinear mapping;

[0131] E2. Calculate the fitness value of the gray wolf individual and the position of the elite individual. The calculation formula for the position of the elite individual is as follows:

[0132] ;

[0133] in, Indicates The iteration The reverse solution of the gray wolf individual, Indicates The iteration The elite solution for the gray wolf individual, Representation interval The random number inside Indicates The maximum position of the gray wolf individual at the iteration, Indicates The minimum position of the individual gray wolf at the iteration;

[0134] E3, marking a preset number of gray wolf individuals with the best fitness values ​​as leader wolves, and updating the positions of the remaining gray wolf individuals in the gray wolf population;

[0135] E4. Calculate and sort the updated individual fitness values ​​of the gray wolves, obtain the elite gray wolves, and update the positions of the elite gray wolves;

[0136] E5. Repeat the above steps until the optimal solution is found or the maximum number of iterations of the gray wolf optimization algorithm is reached, so as to obtain the optimal penalty factor and kernel function parameters of the support vector machine model, and improve the support vector machine model based on the optimal penalty factor and kernel function parameters.

[0137] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A wide-area monitoring method for subsynchronous oscillation of a new energy power grid based on the Internet of Things, characterized in that: include: A wide area monitoring substation is set at each node in the target new energy grid, and a plurality of the wide area monitoring substations communicate data with a wide area monitoring master station, and the wide area monitoring substations collect phasor data of the nodes in the target new energy grid through a PMU to obtain the phasor data; Acquire a network topology diagram of the target new energy power grid, and map the phasor data of each node of the target new energy power grid with the network topology diagram in real time to obtain a phasor flow network topology diagram of the target new energy power grid; The wide area monitoring substation compresses the phasor data to obtain standard phasor data of nodes in the target new energy grid, and uploads the standard phasor data of nodes in the target new energy grid to the wide area monitoring main station; The wide area monitoring master station calculates a unified oscillation frequency of nodes in the target new energy grid based on standard phasor data of the nodes in the target new energy grid, and compares the unified oscillation frequency with a preset frequency threshold; If the error between the two exceeds a preset frequency range, parameter identification is performed on the standard phasor data of the node in the target new energy grid to obtain identification parameters of the node in the target new energy grid, and the identification parameters are compared with a preset parameter threshold; If the parameter error between the two exceeds the preset parameter range, feature extraction is performed on the standard phasor data to obtain a feature set corresponding to the standard phasor data, the feature set is input into a trained state classification model, and the state of the node in the target new energy power grid is output through the trained state classification model; The wide area monitoring substation compresses the phasor data to obtain standard phasor data of nodes in the target new energy power grid, including: Inputting the phasor data into a trained generation model, and compressing the phasor data through the trained generation model to obtain standard phasor data of the target new energy power grid; The generation model adopts a neural network, and the generation model includes an input layer, a hidden layer and an output layer, the input layer is used to input the phasor data, the hidden layer is used to extract hidden features of the phasor data, and the output layer is used to output reconstructed data corresponding to the phasor data based on the hidden features, and the generation model adopts a logistic activation function, if the reconstructed data output by the generation model is different from the phasor data, the generation model is optimized by a loss function; The loss function is as follows: ； in, represents the loss function, Indicates the total length of the data window, The data window represents the phasor data of the node in the target new energy grid The input signal, Represents the data window in the reconstructed data The output signal, represents the weighted regularization term, represents the sparse regularization term, represents the weight coefficient of the sparse regularization term, represents the weight coefficient of the weighted regularization term; The calculation formula of the unified oscillation frequency of the nodes in the target new energy grid is as follows: ； in, Indicates the first The uniform oscillation frequency of the nodes, Represents the total number of nodes in the target new energy grid, Indicates the first The modal voltage amplitude measured at each node, Indicates Nodes and The line current between the nodes, represents the weighted current weight, Indicates the set threshold value. Indicates the first The oscillation frequency measured by each node; Performing parameter identification on standard phasor data of nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid includes: The standard phasor data of the nodes in the target new energy grid are subjected to continuous wavelet transform to obtain the time series data corresponding to the standard phasor data of the nodes in the target new energy grid. The continuous wavelet transform formula is as follows: ； in, represents the continuous wavelet transform formula, Represents standard phasor data, represents the scale factor, represents the translation factor, represents the conjugate wavelet basis; The time series data corresponding to the standard phasor data of the nodes in the target new energy grid is divided into a plurality of continuous frequency intervals, and the frequency interval expression is as follows: ； in, represents the frequency interval expression, Indicates the time series data corresponding to The frequency of The discrete wavelet transform quantities of the plurality of continuous frequency intervals are calculated, and the discrete wavelet transform quantity calculation formula is as follows: ； in, represents the discrete wavelet transform quantity, represents the change in frequency, Indicates the change in scale factor; The time series data is reconstructed based on the discrete wavelet transform amount, and the reconstruction formula of the time series data is as follows: ； in, Represents the time series data IMT components, express Discrete values ​​of ; Performing parameter identification on the standard phasor data of the nodes in the target new energy grid to obtain identification parameters of the nodes in the target new energy grid, further comprising: The standard phasor data is modeled based on the reconstructed time series data to obtain a phasor basic model. The phasor basic model expression is as follows: ； in, Represents the basic phasor model, represents the amplitude, represents the attenuation factor, represents the oscillation frequency, represents the damping ratio, represents the phase angle, represents the undamped frequency; Solving identification parameters of nodes in the target new energy grid based on the phasor basic model to obtain damping ratios and attenuation factors of nodes in the target new energy grid; Performing feature extraction on the standard phasor data to obtain a feature set corresponding to the standard phasor data includes: The amplitude ratio of the standard phasor data is feature extracted by a first formula to obtain a first feature corresponding to the standard phasor data. The first formula is as follows: ； in, represents the amplitude ratio of the standard phasor data, Indicates The amplitude ratio of the frequency components, It represents the ratio of the amplitude of the maximum frequency component to the amplitude of the fundamental component. Indicates the number of frequency components in standard phasor data; The envelope relative amplitude of the standard phasor data is feature extracted by a second formula to obtain a second feature corresponding to the standard phasor data. The second formula is as follows: ； in, Represents the relative amplitude of the envelope of the standard phasor data, Represents the average value of each point in the upper envelope maximum point sequence, represents the average value of each point in the sequence of upper envelope minimum points, Represents the average value of each point on the upper envelope; The envelope extreme value points of the standard phasor data are feature extracted by a third formula to obtain a third feature corresponding to the standard phasor data. The third formula is as follows: ； in, Indicates the number of envelope extreme points of standard phasor data, Indicates the length of the upper envelope maximum point sequence, Indicates the length of the sequence of minimum points of the upper envelope; The first feature, the second feature and the third feature are combined to obtain a feature set corresponding to the standard phasor data.

2. According to claim 1, a method for monitoring subsynchronous oscillation of a new energy power grid based on the Internet of Things is characterized by: Mapping the phasor data of each node of the target new energy grid with the network topology diagram in real time to obtain the phasor flow network topology diagram of the target new energy grid includes: Using a power system modeling tool to graphically model the network topology of the target new energy power grid to obtain a network topology model; Creating a node mapping table to map the plurality of wide area monitoring substations to corresponding nodes in the network topology model; Based on the node mapping table, mapping the phasor data of each node of the target new energy power grid with the corresponding node in the network topology model; The voltage amplitude, phase angle and current flow direction of each node in the network topology model are represented based on color coding, size changes of voltage amplitude symbols and arrows indicating current flow direction to obtain a phasor flow network topology diagram of the target new energy power grid.

3. According to a method for monitoring subsynchronous oscillation of a new energy power grid based on the Internet of Things in claim 1, it is characterized in that: The state classification model adopts an improved support vector machine model, and the state classification model improves the support vector machine model through a grey wolf optimization algorithm.

4. According to claim 3, a method for monitoring subsynchronous oscillation of a new energy power grid based on the Internet of Things is characterized in that: The support vector machine model is improved by using the Grey Wolf Optimization Algorithm, including: The initialization parameters of the gray wolf optimization algorithm are set to initialize the position of the gray wolf population. The gray wolf population position initialization formula is as follows: ； in, and Indicates The iteration and The position of the gray wolf individual at the iteration, Represents the parameters of the preset control nonlinear mapping; Calculate the fitness value of the gray wolf individual and calculate the elite individual position. The elite individual position calculation formula is as follows: ； in, Indicates The first iteration The reverse solution of the gray wolf individual, Indicates The first iteration The elite solution for the gray wolf individual, Representation interval The random number inside Indicates The maximum position of the gray wolf individual at the iteration, Indicates The minimum position of the individual gray wolf at the iteration; Marking a preset number of the gray wolf individuals with the best fitness values ​​as leader wolves, and updating the positions of the remaining gray wolf individuals in the gray wolf population; Calculate the updated individual fitness values ​​of the gray wolves and sort them, obtain the elite gray wolves, and update the positions of the elite gray wolves; Repeat the above steps until the optimal solution is found or the maximum number of iterations of the gray wolf optimization algorithm is reached, so as to obtain the optimal penalty factor and kernel function parameters of the support vector machine model, and improve the support vector machine model based on the optimal penalty factor and kernel function parameters.

Images