Multi-source fusion smart power grid harmonic dynamic management system based on harmonic impedance

Abstract

The invention discloses a multi-source fusion intelligent power grid harmonic dynamic management system based on harmonic impedance, and relates to the technical field of intelligent power grids, and the system comprises a multi-source data collection module which collects harmonic impedance, power grid operation, weather and user power consumption behavior data by means of an intelligent electric meter and sensor equipment, the multi-source fusion intelligent power grid harmonic dynamic management system based on harmonic impedance is constructed, comprehensive and accurate management of the power grid harmonic problem is achieved, and the system can automatically collect and process a large amount of data including harmonic impedance, power grid operation, weather and user power consumption behavior data. The method provides powerful support for identification and analysis of harmonic problems, and the system can generate a more accurate and comprehensive comprehensive data set through multi-source deep fusion of the data, thereby realizing accurate positioning and tracking of harmonic sources and deep analysis of harmonic propagation paths, improving the efficiency and accuracy of power grid harmonic management, and reducing the cost of power grid harmonic management. And a powerful guarantee is provided for safe and stable operation of a power grid.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance. Background Art

[0002] With the rapid development of smart grids and the widespread application of power electronic equipment, the large-scale construction of AC transmission networks above 750 kV and the establishment of large-scale grid security and defense systems have placed higher demands on grid harmonic management. In the coordinated operation of intelligent dispatching systems, harmonics have become a key factor affecting the stability of power supply. Harmonics not only lead to a decline in power quality but can also cause equipment overheating and relay protection malfunctions, seriously threatening the safe operation of transmission lines above 750 kV. Traditional harmonic management technologies have the following main shortcomings:

[0003] Single data collection dimension

[0004] The existing system is unable to fully cover the harmonic impedance characteristic monitoring of transmission lines above 750 kV, and is insufficient in its ability to analyze the synergistic impact of distributed energy and nonlinear loads in large-scale power grids, and cannot meet the intelligent dispatching system's demand for data at multiple time and space scales.

[0005] Limited precision of analytical methods

[0006] Traditional positioning algorithms rely on single-point measurements and are unable to adapt to the tracking of harmonic propagation paths under the complex topology of large-scale power grids. Especially in power grids above 750 kV, the harmonic reflection and superposition effects are significant, making it difficult for traditional methods to accurately identify the dynamics of harmonic sources.

[0007] Lack of coordination in management decision-making

[0008] Existing solutions are not deeply coupled with intelligent dispatch systems and are unable to respond in real time to the impact of changes in grid operation (such as power flow shifts and equipment commissioning and decommissioning) on harmonic distribution. This results in insufficient coordinated optimization of governance measures and large-scale grid security defense systems. Therefore, developing a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance will significantly improve grid harmonic management and ensure the safe and stable operation of smart grids. Summary of the Invention

[0009] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance. The system integrates multiple functional modules such as multi-source data acquisition, deep data fusion, harmonic source positioning and tracking, harmonic propagation path analysis, and comprehensive evaluation and management decision-making. Through intelligent means, it realizes comprehensive monitoring, precise analysis and scientific management of power grid harmonic problems. It can not only accurately identify and locate harmonic sources, but also effectively evaluate the harmonic risk level, and formulate targeted harmonic control plans based on the evaluation results.

[0010] In order to solve the above technical problems, the present invention provides the following technical solutions: a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance, the system comprising:

[0011] Multi-source data acquisition module: This module uses smart meters and sensor devices to collect data on harmonic impedance, grid operation, weather conditions, and user electricity usage behavior. The data is cleaned and normalized locally through embedded edge storage devices and transmitted to the multi-source data deep fusion module.

[0012] Multi-source data deep fusion module: Receives data transmitted from the multi-source data acquisition module, uses principal component analysis to reduce the data dimension, and uses the multi-source data fusion algorithm to fuse different types of data to generate a comprehensive data set;

[0013] Harmonic source positioning and tracking module: Receives the comprehensive data set output by the data preprocessing and fusion module, calculates the harmonic source position coordinates based on the harmonic impedance characteristics, and uses a positioning algorithm based on the harmonic impedance gradient. It also monitors and continuously tracks the harmonic source dynamics in real time.

[0014] Harmonic Propagation Path Analysis Module: This module obtains grid topology data and harmonic source information provided by the harmonic source location and tracking module, constructs a graph model of the grid using grid nodes and lines, calculates complex power flows based on node voltages and line electrical parameters, determines the propagation direction of harmonics based on the phase of the complex power flows, and determines the propagation intensity of harmonics on different lines by calculating the propagation intensity coefficient.

[0015] Comprehensive assessment and management decision-making module: Based on the outputs of the multi-source data deep fusion module, the harmonic source location and tracking module, and the harmonic propagation path analysis module, the long short-term memory network is used to predict the probability of equipment failure caused by harmonics, and a harmonic management plan is formulated based on the assessment results.

[0016] Furthermore, the data types and acquisition devices collected in the multi-source data acquisition module are:

[0017] Harmonic impedance data:

[0018] Data type: The amplitude and phase relationship of harmonic voltage and harmonic current in the power grid and the harmonic impedance value at a specific frequency;

[0019] Acquisition equipment: harmonic analyzer and harmonic impedance measurement device;

[0020] Grid operation data:

[0021] Data type: voltage, current, power and frequency;

[0022] Collection equipment: smart meters, voltage transformers, current transformers and synchronized phasor measurement devices Meteorological data:

[0023] Data type: temperature, humidity, wind speed and air pressure;

[0024] Collection equipment: temperature sensor, humidity sensor, anemometer and barometer;

[0025] User electricity usage behavior data:

[0026] Data types: electricity usage time, electricity power, appliance usage patterns and abnormal electricity usage behavior,

[0027] Collection equipment: smart meters, non-intrusive load monitoring devices used to identify user electrical equipment through current waveform analysis, and user-side power quality monitoring sensors.

[0028] Furthermore, the multi-source data deep fusion module uses principal component analysis to reduce the dimension of the data, receives the data pre-processed by the multi-source data acquisition module, and constructs the covariance matrix C of the data. The calculation formula is: in is the ith eigenvalue of the kth sample after standardization, n is the number of samples, is the mean of the i-th feature after normalization, is the jth sample of the kth sample after normalization, Represents the mean of the jth feature after standardization, and performs eigenvalue decomposition on the covariance matrix C to obtain the eigenvalue λ i and the corresponding eigenvector e i , i=1,2,…,p, p is the number of original features, λ i The i-th eigenvalue of the covariance matrix C represents the variance of the data in that direction, e i Denotes the covariance matrix C corresponding to the eigenvalue λ i The eigenvector of represents the direction of data change. The eigenvalues are arranged in descending order. According to the size of the eigenvalue, the top m principal components with a cumulative contribution rate of 85%-95% are retained. The calculation formula is: Where m is the number of principal components selected, and the eigenvectors corresponding to the first m principal components selected form the transformation matrix to convert the data matrix X * Multiply by the transformation matrix E, X * The data matrix after normalization is the data matrix Y after dimensionality reduction, and the formula is: Y = X * E, E is the transformation matrix composed of the eigenvectors corresponding to the first m principal components selected.

[0029] Furthermore, the multi-source data deep fusion module fuses different types of data with the help of multi-source data fusion algorithm. Let the i-th type of data collected be D i , whose adaptive weight is w i , the fused data is D f , the calculation formula is: Where c is the number of data samples and o is the number of data types.

[0030] Furthermore, the harmonic source positioning and tracking module uses a positioning algorithm based on harmonic impedance gradient to calculate the harmonic source position coordinates. Suppose the harmonic impedance at a point x in the power grid is Z(x), the harmonic source position is x0, and x is calculated by iteration. k+1 To approximate the harmonic source position x0, the calculation formula is: in is x k The gradient of the harmonic impedance at x k is the point that approaches the harmonic source position at the kth iteration, x k+1 The point approaching the harmonic source position in the k+1th iteration, α is the step size parameter, and the value range is 0.1-0.5. The value range is 0.1-0.5. α is dynamically adjusted according to the fluctuation of the real-time harmonic impedance gradient. The adjustment formula is: Among them: α0 is the initial step length coefficient, and the calculation formula is: The harmonic impedance gradient at the initial iteration point, η is the proportional coefficient, ranging from 0.2 to 0.8, and δ is the smoothing constant to prevent the denominator from failing when the gradient is zero. is the harmonic impedance gradient at the kth iteration, λ is the sensitivity adjustment factor, and the recommended value is 0.5 to 1.5, and ∈ is a very small positive number to avoid the denominator being zero.

[0031] Furthermore, the harmonic propagation path analysis module uses the grid nodes and lines to abstractly construct a graph structure model of the grid. The obtained grid topology data is used to abstract the nodes in the grid as vertices of the graph and the lines as edges connecting the vertices. The graph structure model of the grid is constructed G = (V, E), where α is the node set, E is the edge set, and each node v i ∈V corresponds to an actual node in the power grid, and each edge eij ∈E connects to node v i and v j , representing the actual transmission line, and assigning electrical parameters to each edge.

[0032] Furthermore, the harmonic propagation direction is determined in the harmonic propagation path analysis module. For any two nodes i and j connected by a line in the power grid, the complex power flow S between them is calculated. ij , the calculation formula is: Where V i is the voltage phasor at node i, obtained by the voltage measuring device installed at node i, isI ij The complex conjugate of ij is the current phasor flowing from node i to node j, I ij The voltage V across node i i , the voltage V at node j j And the electrical parameters of the circuit are calculated. For the circuit including the resistor R ij 、Inductor L ij and capacitor C ij The line current phasor I ij The calculation formula is: Where ω is the angular frequency, π = 2πf h , f h is the frequency of the hth harmonic, calculated as: f h =h·f0, where h is the harmonic order, f0 is the grid fundamental frequency, which is 50Hz, and the propagation direction of the harmonic is determined by the complex power flow S ij The phase of Re(S ij )>0, Re represents the real part, indicating that the power flows from node i to node j, that is, the harmonic propagates from node i to node j along the line; when Re(S ij )>0, the harmonic propagation direction is opposite.

[0033] Furthermore, the harmonic propagation path analysis module determines the propagation intensity of harmonics on different lines by calculating the propagation intensity coefficient. Let the propagation intensity coefficient be K ij , the calculation formula is: where |S ij | is the complex power flow S ij The modulus indicates the magnitude of the harmonic power on the line, N i is the set of adjacent nodes of node i, represents the sum of the complex power flow modules between node i and all adjacent nodes.

[0034] Furthermore, the comprehensive assessment and management decision module uses a long short-term memory network to predict the probability of equipment failure caused by harmonics. From the output of the multi-source data deep fusion module, the harmonic source location tracking module, and the harmonic propagation path analysis module, data related to harmonics and equipment status are extracted, the data is cleaned and standardized, and the data is arranged in chronological order to construct time series data. An LSTM model is constructed, and the number of layers and the number of neurons in each layer are determined. The LSTM unit contains an input gate f t , where x t is the current input, h t-I is the hidden state of the previous moment, W ii and W hi is the weight matrix, b i is the bias vector, σ is the sigmoid activation function, Forget Gate f t :f t =σ(W if x t +W hf h t-1 +b f ), candidate memory units Where tanh is the hyperbolic tangent activation function, Memory unit C t : Where ⊙ represents element-by-element multiplication, and the output gate o t :o t =σ(W i oxt+W h oh t-1 +b o ), hidden state h t :h t =o t ⊙tanh(C t ), add a fully connected layer, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, and then a fully connected layer is added, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b outis the bias vector, the mean square error is selected as the loss function, the Adam optimizer is used to update the parameters, and the preprocessed data is divided into training set, validation set and test set. Assume that the total amount of data is N, the training set ratio is p1, the validation set ratio is p2, the test set ratio is p3, and p1+p2+p3=I, then the number of training sets N1=p1N, the number of validation sets N2=p2N, and the number of test sets N3=p3N. Input the training set data into the LSTM model for training. In each training step, the predicted value is calculated by forward propagation. Then, the loss L is calculated according to the loss function, and the gradient is calculated through the back-propagation algorithm. Finally, the optimizer is used to update the model parameters. The model is optimized through the validation set. The multi-source data collected in real time is processed according to the data preprocessing steps and input into the trained LSTM model. The model outputs the predicted equipment failure probability P fault , when P fault When T1 < P, it is judged as a low risk level, indicating that the harmonic condition in the power grid is relatively good; when T1 < P fault When ≤T2, it is judged as medium risk level, and the power grid has a certain degree of harmonic problem; when When , it is judged as a high risk level, indicating that the power grid harmonic pollution is relatively serious, where T1 and T2 are thresholds set based on industry standards.

[0035] Furthermore, the comprehensive assessment and management decision module formulates a harmonic management plan based on the assessment results:

[0036] Low risk level: Adopt a regular monitoring strategy to continuously track the grid harmonics status and perform preventive maintenance on equipment.

[0037] Medium risk level: In addition to strengthening frequency monitoring, further analyze harmonic sources to identify the main harmonic-generating equipment or areas, and take measures such as adjusting load distribution, optimizing transformer taps, and installing small harmonic control devices;

[0038] High-risk level: Immediately initiate emergency control measures, shut down and rectify harmonic source equipment that seriously exceeds the standard, replace large-capacity, high-performance active power filters and static VAR compensators, and install harmonic isolation devices on key equipment that is seriously affected by harmonics.

[0039] Compared with the existing technology, a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance has the following beneficial effects:

[0040] 1. The present invention realizes comprehensive and accurate management of power grid harmonic problems by constructing a multi-source fusion intelligent grid harmonic dynamic management system based on harmonic impedance. The system can automatically collect and process a large amount of data, including harmonic impedance, power grid operation, meteorological and user electricity consumption behavior data, which provides strong support for the identification and analysis of harmonic problems. Through the deep fusion of multi-source data, the system can generate a more accurate and comprehensive integrated data set, thereby realizing the precise positioning and tracking of harmonic sources and in-depth analysis of harmonic propagation paths, improving the efficiency and accuracy of power grid harmonic management, and providing a strong guarantee for the safe and stable operation of the power grid.

[0041] 2. The present invention uses a positioning algorithm based on harmonic impedance gradient to quickly and accurately calculate the position coordinates of harmonic sources, thereby realizing real-time monitoring and continuous tracking of harmonic sources. At the same time, by abstracting the grid nodes and lines to construct a graph structure model of the grid, combined with the calculation of complex power flow and the determination of the harmonic propagation intensity coefficient, the system can comprehensively evaluate the propagation of harmonics in the grid and its potential impact, which not only improves the ability to identify and analyze harmonic problems, but also provides a strong basis for formulating scientific and reasonable harmonic management plans. By implementing targeted governance measures, the level of harmonic pollution in the grid can be significantly reduced, and the grid operation efficiency and power quality can be improved.

[0042] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0044] Figure 1 This is a flow chart of a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance;

[0045] Figure 2 This is a framework diagram of a multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance. DETAILED DESCRIPTION

[0046] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0047] Example 1:

[0048] Smart grid harmonic management in urban commercial areas.

[0049] In urban commercial areas, numerous commercial electrical equipment, office equipment and lighting systems cause the power grid to face complex harmonic problems.

[0050] Data collection: Smart meters, harmonic analyzers, voltage transformers, current transformers, and synchronized phasor measurement devices are installed at distribution boxes, transformers, and major electrical equipment in commercial areas. Smart meters collect user electricity usage data such as electricity usage time, power, and appliance usage patterns, as well as voltage, current, power, and frequency data. Harmonic analyzers collect the amplitude and phase relationship of harmonic voltage and current, as well as harmonic impedance values at specific frequencies. Temperature sensors, humidity sensors, anemometers, and barometers collect meteorological data. These data are collected by the multi-source data acquisition module, cleaned and normalized locally through embedded edge storage devices, and then transmitted to the multi-source data deep fusion module.

[0051] Data fusion and dimensionality reduction: The multi-source data deep fusion module receives the pre-processed data and constructs the data covariance matrix C. The calculation formula is: in is the ith eigenvalue of the kth sample after standardization, n is the number of samples, is the mean of the i-th feature after normalization is the jth value of the kth sample after normalization Represents the mean of the jth feature after standardization, and performs eigenvalue decomposition on the covariance matrix C to obtain the eigenvalue λ i and the corresponding eigenvector e i , i=1,2,…,p, p is the number of original features, λ i The i-th eigenvalue of the covariance matrix C represents the variance of the data in that direction. The eigenvalues are sorted from large to small. Based on the size of the eigenvalues, the top m principal components with a cumulative contribution rate of 85%-95% are retained. The cumulative contribution rate is calculated as follows: Where m is the number of principal components selected, and the eigenvectors corresponding to the first m principal components selected form the transformation matrix to convert the data matrix X * Multiply by the transformation matrix E, X * The data matrix after normalization is the data matrix Y after dimensionality reduction, that is, Y = X * E, then, using the multi-source data fusion algorithm, let the collected i-th type of data be D i , whose adaptive weight is w i , the fused data is D f , the calculation formula is: Where c is the number of data samples and o is the number of data types, so that different types of data can be integrated into a comprehensive data set.

[0052] Harmonic source location tracking: The harmonic source location tracking module receives a comprehensive data set and uses a location algorithm based on harmonic impedance gradient to calculate the harmonic source location coordinates. Suppose the harmonic impedance at a point x in the power grid is Z(x), the harmonic source location is x0, and x is calculated iteratively. k+1 To approximate the harmonic source position x0, the calculation formula is: in is x k The gradient of the harmonic impedance at x k is the point that approaches the harmonic source position at the kth iteration, x k+1 The point approaching the harmonic source position in the k+1th iteration, a is the step size parameter, and the value range is 0.1-0.5. a is dynamically adjusted according to the fluctuation of the real-time harmonic impedance gradient. The adjustment formula is: Among them: α0 is the initial step length coefficient, and the calculation formula is: The harmonic impedance gradient at the initial iteration point, η is the proportional coefficient, ranging from 0.2 to 0.8, and δ is the smoothing constant to prevent the denominator from failing when the gradient is zero. is the harmonic impedance gradient at the kth iteration, λ is the sensitivity adjustment factor, and the recommended value is 0.5 to 1.5, and ∈ is a very small positive number to avoid the denominator being zero.

[0053] Harmonic propagation path analysis: Obtain the topological structure data of the commercial area power grid, abstract the nodes in the power grid into graph vertices, and abstract the lines into edges connecting the vertices. Construct a graph structure model G = (V, E), where V is the node set and E is the edge set. Each node v i ∈V corresponds to an actual node in the power grid, and each edge e ij ∈E connects to node v i and v j , represents the actual transmission line, assigns electrical parameters to each edge, and takes the transmission line connecting the shopping mall and the nearby office building as an example to calculate the complex power flow S between them ij , the calculation formula of complex power flow is: Where V i is the voltage phasor at node i, obtained by the voltage measuring device installed at node i, isI ij The complex conjugate of ij is the current phasor flowing from node i to node j. According to circuit theory, I ij The voltage V across node i i , the voltage V at node j j And the electrical parameters of the circuit are calculated. For the circuit including the resistor Rij 、Inductor L ij and capacitor C ij The line current phasor I ij The calculation formula is: Where ω is the angular frequency, ω = 2πf h , f h is the frequency of the hth harmonic, calculated as: f h =h·f0, h is the harmonic order, f0 is the fundamental frequency of the power grid, according to the complex power flow S ij The phase of the harmonic propagation direction is determined by the phase of the harmonic propagation direction. When Re(S ij )>0, Re represents the real part, indicating that the power flows from node i to node j, that is, the harmonic propagates from node i to node j along the line; when Re(S ij )>0, the harmonic propagation direction is opposite, and the propagation intensity of the harmonic on different lines is determined by calculating the propagation intensity coefficient. Let the propagation intensity coefficient be K ij , the calculation formula is: where |S ij | is the complex power flow S ij The modulus indicates the magnitude of the harmonic power on the line, N i is the set of adjacent nodes of node i, It represents the sum of the complex power flow modules between node i and all adjacent nodes, and determines the proportion of the harmonic propagation intensity on the line in the entire power grid, where Ni is the set of adjacent nodes of node i.

[0054] Comprehensive evaluation and management decision-making: The comprehensive evaluation and management decision-making module uses the long short-term memory network to predict the probability of equipment failure caused by harmonics based on the output results of the previous modules. It extracts data related to harmonics and equipment status from the output of the multi-source data deep fusion module, the harmonic source location and tracking module, and the harmonic propagation path analysis module. It cleans and standardizes the data and arranges the data in chronological order to construct time series data. It also constructs an LSTM model and determines the number of layers and the number of neurons in each layer. The LSTM unit contains an input gate f t , where x t is the current input, h t-I is the hidden state of the previous moment, W ii and W hi is the weight matrix, b i is the bias vector, σ is the sigmoid activation function, Forget Gate f t :f t =σ(W if x t +W hf h t-1 +b f), candidate memory units Where tanh is the hyperbolic tangent activation function, Memory unit C t : Where ⊙ represents element-by-element multiplication, and the output gate o t :o t =σ(W i oxt+W h oh t-1 +b o ), hidden state h t :h t =o t ⊙tanh(C t ), add a fully connected layer, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, and then a fully connected layer is added, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, the mean square error is selected as the loss function, the Adam optimizer is used to update the parameters, and the preprocessed data is divided into training set, validation set and test set. Assume that the total amount of data is N, the training set ratio is p1, the validation set ratio is p2, the test set ratio is p3, and p1+p2+p3=1, then the number of training sets N1=p1N, the number of validation sets N2=p2N, and the number of test sets N3=p3N. Input the training set data into the LSTM model for training. In each training step, the predicted value is calculated by forward propagation. Then, the loss L is calculated according to the loss function, and the gradient is calculated through the back-propagation algorithm. Finally, the optimizer is used to update the model parameters. The model is optimized through the validation set. The multi-source data collected in real time is processed according to the data preprocessing steps and input into the trained LSTM model. The model outputs the predicted equipment failure probability P fault , after calculation, T1<Y iThe assessment result of ≤T2 is a medium risk level. In addition to strengthening the monitoring frequency, further analysis of the harmonic source revealed that the frequency conversion control systems of some old elevators in the mall are the main harmonic generating devices. Measures such as adjusting the elevator running time, optimizing the transformer taps, and installing small harmonic control devices in the elevator distribution boxes have been taken to reduce the impact of harmonics on the power grid.

[0055] In summary, in the smart grid harmonic management of urban commercial areas, the multi-source data acquisition module is used to comprehensively collect various types of data, and the multi-source data deep fusion module is used to perform dimensionality reduction and fusion processing to provide strong support for subsequent analysis. The harmonic source positioning and tracking module accurately locks the harmonic source, and the harmonic propagation path analysis module clarifies the direction and intensity of harmonic propagation. The comprehensive assessment and management decision-making module assesses risks based on the results of each module and takes effective measures for medium-risk levels. This series of operations constitutes a complete harmonic management system that can promptly discover and solve harmonic problems in commercial areas, ensure stable operation of the power grid, improve power quality, and ensure the normal operation of various electrical equipment.

[0056] Example 2:

[0057] Smart grid harmonic management in industrial parks.

[0058] A large number of industrial production equipment in industrial parks, such as motors, welding machines, and rectifiers, will generate a large number of harmonics, threatening the stability of the power grid.

[0059] Data collection: Smart meters, harmonic analyzers, voltage transformers, current transformers, and synchronized phasor measurement devices are installed in the industrial park's power distribution rooms, production workshop distribution boxes, and major production equipment to collect data on electricity consumption, grid operation, and harmonic impedance. Meteorological data is collected using meteorological sensors within the park. This data is cleaned and normalized locally through embedded edge storage devices and then transmitted to the multi-source data deep fusion module.

[0060] Data fusion and dimensionality reduction: The multi-source data deep fusion module constructs the covariance matrix C, which is calculated as follows: And perform eigenvalue decomposition to obtain the eigenvalue λ i and the corresponding eigenvector e i , sort by eigenvalue size, and retain the top m principal components whose cumulative contribution rate reaches 85%-95% based on the size of the eigenvalue. The calculation formula for the cumulative contribution rate is: The eigenvectors corresponding to the first m principal components are selected to form a transformation matrix to convert the data matrix X * Multiply by the transformation matrix E, X * The data matrix after normalization is the data matrix Y after dimensionality reduction, that is, Y = X * E. Use multi-source data fusion algorithm to fuse different types of data. Let the collected i-th type of data be Di , whose adaptive weight is w i , the fused data is D f , the calculation formula is: Merge different types of data into comprehensive datasets.

[0061] Harmonic source location tracking: The harmonic source location tracking module receives a comprehensive data set and uses a location algorithm based on harmonic impedance gradient to calculate the harmonic source location coordinates. Suppose the harmonic impedance at a point x in the power grid is Z(x), the harmonic source location is x0, and x is calculated iteratively. k+1 To approximate the harmonic source position x0, the calculation formula is: Calculations revealed that a large electric welding machine in a metal processing plant within the park was the main source of harmonics, and its operating status was tracked in real time.

[0062] Harmonic propagation path analysis: Obtain the industrial park power grid topology data, construct a graph structure model G = (V, E), assign electrical parameters to the edges, and use the transmission line connecting the metal processing plant and the adjacent electronics factory as an example to calculate the complex power flow formula: The harmonic propagation direction is determined based on the phase of the complex power flow, and the propagation intensity of the harmonics on different lines is determined by calculating the propagation intensity coefficient. Let the propagation intensity coefficient be K ij , the calculation formula is:

[0063] Comprehensive evaluation and management decision-making: The comprehensive evaluation and management decision-making module uses the long short-term memory network to predict the probability of equipment failure caused by harmonics based on the output of each module. It extracts data related to harmonics and equipment status from the output of the multi-source data deep fusion module, the harmonic source location tracking module, and the harmonic propagation path analysis module. It cleans and standardizes the data and arranges the data in chronological order to construct time series data. It also constructs an LSTM model and determines the number of layers and the number of neurons in each layer. The LSTM unit contains an input gate f t , where x t is the current input, h t-1 is the hidden state of the previous moment, W ii and W hi is the weight matrix, b i is the bias vector, σ is the sigmoid activation function, Forget Gate f t :f t =σ(W if x t +W hf h t-1 +b f ), candidate memory units Where tanh is the hyperbolic tangent activation function, Memory unit C t : Where ⊙ represents element-by-element multiplication, and the output gate ot ： o t =σ(W i oxt+W h oh t-1 +b o ), hidden state h t :h t =o t ⊙tanh(C t ), add a fully connected layer, whose output is: y = W out h last +b out , where W old is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, and then a fully connected layer is added, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, the mean square error is selected as the loss function, the Adam optimizer is used to update the parameters, and the preprocessed data is divided into training set, validation set and test set. Assume that the total amount of data is N, the training set ratio is p1, the validation set ratio is p2, the test set ratio is p3, and p1+p2+p3=1, then the number of training sets N1=p1N, the number of validation sets N2=p2N, and the number of test sets N3=p3N. Input the training set data into the LSTM model for training. In each training step, the predicted value is calculated by forward propagation. Then, the loss L is calculated according to the loss function, and the gradient is calculated through the back-propagation algorithm. Finally, the optimizer is used to update the model parameters. The model is optimized through the validation set. The multi-source data collected in real time is processed according to the data preprocessing steps and input into the trained LSTM model. The model outputs the predicted equipment failure probability P fault After calculation and assessment, the result was a high risk level, and emergency treatment measures were immediately initiated. The electric welding machines in the metal processing plant were shut down for rectification, and large-capacity, high-performance active power filters were replaced. Harmonic isolation devices were installed in front of key production equipment in the electronics factory that was seriously affected by harmonics to ensure the stable operation of the power grid and the normal operation of other equipment.

[0064] In summary, the smart grid harmonic management embodiment of the industrial park demonstrates the application of the present invention in a complex industrial power environment. The multi-source data acquisition equipment collects rich data in real time, the multi-source data deep fusion module optimizes the data processing, the harmonic source positioning and tracking module accurately finds the harmonic source, the harmonic propagation path analysis module grasps the harmonic propagation situation, and the comprehensive evaluation and management decision-making module decisively implements emergency control measures at high risk levels based on the evaluation results. The system effectively responds to the serious problem of harmonic pollution in the industrial park, reduces the harm of harmonics to the power grid and equipment, ensures the continuity and stability of industrial production, and improves the overall power supply reliability of the industrial park.

[0065] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance, characterized in that: The system includes: Multi-source data acquisition module: This module uses smart meters and sensor devices to collect data on harmonic impedance, grid operation, weather conditions, and user electricity usage behavior. The data is cleaned and normalized locally through embedded edge storage devices and transmitted to the multi-source data deep fusion module. Multi-source data deep fusion module: Receives data transmitted from the multi-source data acquisition module, uses principal component analysis to reduce the data dimension, and uses the multi-source data fusion algorithm to fuse different types of data to generate a comprehensive data set; Harmonic source positioning and tracking module: Receives the comprehensive data set output by the data preprocessing and fusion module, calculates the harmonic source position coordinates based on the harmonic impedance characteristics, and uses a positioning algorithm based on the harmonic impedance gradient. It also monitors and continuously tracks the harmonic source dynamics in real time. Harmonic Propagation Path Analysis Module: This module obtains grid topology data and harmonic source information provided by the harmonic source location and tracking module, constructs a graph model of the grid using grid nodes and lines, calculates complex power flows based on node voltages and line electrical parameters, determines the propagation direction of harmonics based on the phase of the complex power flows, and determines the propagation intensity of harmonics on different lines by calculating the propagation intensity coefficient. Comprehensive assessment and management decision-making module: Based on the outputs of the multi-source data deep fusion module, the harmonic source location and tracking module, and the harmonic propagation path analysis module, the long short-term memory network is used to predict the probability of equipment failure caused by harmonics, and a harmonic management plan is formulated based on the assessment results.

2. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The data types and acquisition devices collected in the multi-source data acquisition module are: Harmonic impedance data: Data type: The amplitude and phase relationship of harmonic voltage and harmonic current in the power grid and the harmonic impedance value at a specific frequency; Acquisition equipment: harmonic analyzer and harmonic impedance measurement device; Grid operation data: Data type: voltage, current, power and frequency; Collection equipment: smart meters, voltage transformers, current transformers and synchronized phasor measurement devices Meteorological data: Data type: temperature, humidity, wind speed and air pressure; Collection equipment: temperature sensor, humidity sensor, anemometer and barometer; User electricity usage behavior data: Data types: electricity usage time, electricity power, appliance usage patterns and abnormal electricity usage behavior, Collection equipment: smart meters, non-intrusive load monitoring devices used to identify user electrical equipment through current waveform analysis, and user-side power quality monitoring sensors.

3. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The multi-source data deep fusion module uses principal component analysis to reduce the data dimension, receives the data pre-processed by the multi-source data acquisition module, and constructs the data covariance matrix C. The calculation formula is: in is the ith eigenvalue of the kth sample after standardization, n is the number of samples, is the mean of the i-th feature after normalization, is the jth sample of the kth sample after normalization, Represents the mean of the jth feature after standardization, and performs eigenvalue decomposition on the covariance matrix C to obtain the eigenvalue λ i and the corresponding eigenvector e i , i=1,2,…,p, p is the number of original features, λ i The i-th eigenvalue of the covariance matrix C represents the variance of the data in that direction, e i Denotes the covariance matrix C corresponding to the eigenvalue λ i The eigenvector of represents the direction of data change. The eigenvalues are arranged in descending order. According to the size of the eigenvalue, the top m principal components with a cumulative contribution rate of 85%-95% are retained. The calculation formula is: Where m is the number of principal components selected, and the eigenvectors corresponding to the first m principal components selected form the transformation matrix to convert the data matrix X * Multiply by the transformation matrix E, X * The data matrix after normalization is the data matrix Y after dimensionality reduction, and the formula is: Y = X * E, E is the transformation matrix composed of the eigenvectors corresponding to the first m principal components selected.

4. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The multi-source data deep fusion module uses the multi-source data fusion algorithm to fuse different types of data. Let the i-th type of data collected be D i , whose adaptive weight is w i , the fused data is D f , the calculation formula is: Where c is the number of data samples and o is the number of data types.

5. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The harmonic source positioning and tracking module uses a positioning algorithm based on harmonic impedance gradient to calculate the harmonic source position coordinates. Suppose the harmonic impedance at a point x in the power grid is Z(x), the harmonic source position is x0, and x is calculated by iteration. k+1 To approximate the harmonic source position x0, the calculation formula is: in is x k The gradient of the harmonic impedance at x k is the point that approaches the harmonic source position at the kth iteration, x k+1 The point approaching the harmonic source position in the k+1th iteration, a is the step size parameter, and the value range is 0.1-0.

5. a is dynamically adjusted according to the fluctuation of the real-time harmonic impedance gradient. The adjustment formula is: Among them: α0 is the initial step length coefficient, and the calculation formula is: The harmonic impedance gradient at the initial iteration point, η is the proportional coefficient, ranging from 0.2 to 0.8, and δ is the smoothing constant to prevent the denominator from failing when the gradient is zero. is the harmonic impedance gradient at the kth iteration, λ is the sensitivity adjustment factor, and the recommended value is 0.5 to 1.5, and ∈ is a very small positive number to avoid the denominator being zero.

6. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The harmonic propagation path analysis module uses the grid nodes and lines to abstractly construct a graph structure model of the grid. The obtained grid topology data is abstracted into the nodes in the grid as the vertices of the graph and the lines as the edges connecting the vertices. The graph structure model G = (V, E) of the grid is constructed, where V is the node set and E is the edge set. Each node v i ∈V corresponds to an actual node in the power grid, and each edge e ij ∈E connects to node v i and v j , representing the actual transmission line, and assigning electrical parameters to each edge.

7. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The harmonic propagation direction in the harmonic propagation path analysis module is determined by calculating the complex power flow S between two nodes i and j connected by any line in the power grid. ij , the calculation formula is: Where V i is the voltage phasor at node i, obtained by the voltage measuring device installed at node i, isI ij The complex conjugate of ij is the current phasor flowing from node i to node j, I ij The voltage V across node i i , the voltage V at node j j And the electrical parameters of the circuit are calculated. For the circuit including the resistor R ij 、Inductor L ij and capacitor C ij The line current phasor I ij The calculation formula is: Where ω is the angular frequency, ω2πf h , f h is the frequency of the hth harmonic, calculated as: f h =h·f0, where h is the harmonic order, f0 is the grid fundamental frequency, which is 50Hz, and the propagation direction of the harmonic is determined by the complex power flow S ij The phase of Re(S ij )>0, Re represents the real part, indicating that the power flows from node i to node j, that is, the harmonic propagates from node i to node j along the line; when Re(S ij )>0, the harmonic propagation direction is opposite.

8. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 7 is characterized in that: The harmonic propagation path analysis module determines the propagation intensity of harmonics on different lines by calculating the propagation intensity coefficient. Let the propagation intensity coefficient be K ij , the calculation formula is: where |S ij | is the complex power flow S ij The modulus indicates the magnitude of the harmonic power on the line, N i is the set of adjacent nodes of node i, represents the sum of the complex power flow modules between node i and all adjacent nodes.

9. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1 is characterized in that: The comprehensive assessment and management decision module uses a long short-term memory network to predict the probability of equipment failure caused by harmonics. From the output of the multi-source data deep fusion module, the harmonic source location tracking module, and the harmonic propagation path analysis module, data related to harmonics and equipment status are extracted, cleaned and standardized, and arranged in chronological order to construct time series data. An LSTM model is constructed, and the number of layers and the number of neurons in each layer are determined. The LSTM unit contains an input gate f t , where x t is the current input, h t-I is the hidden state of the previous moment, W ii and W hi is the weight matrix, b i is the bias vector, σ is the sigmoid activation function, Forget Gate f t :f t =σ(W if x t +W hf h t-1 +b f ), candidate memory units Where tanh is the hyperbolic tangent activation function, Memory unit C t : Where ⊙ represents element-by-element multiplication, and the output gate o t :o t =σ(W i oxt+W h oh t-1 +b o ), hidden state h t :h t =o t ⊙tanh(C t ), add a fully connected layer, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, and then a fully connected layer is added, whose output is: y = W out h last +b out , where W out is the weight matrix of the fully connected layer, h last is the hidden state of the last time step of the last LSTM layer, b out is the bias vector, the mean square error is selected as the loss function, the Adam optimizer is used to update the parameters, and the preprocessed data is divided into training set, validation set and test set. Assume that the total amount of data is N, the training set ratio is p1, the validation set ratio is p2, the test set ratio is p3, and p1+p2+p3=1, then the number of training sets N1=p1N, the number of validation sets N2=p2N, and the number of test sets N3=p3N. Input the training set data into the LSTM model for training. In each training step, the predicted value is calculated by forward propagation. Then, the loss l is calculated according to the loss function, and the gradient is calculated through the back-propagation algorithm. Finally, the optimizer is used to update the model parameters. The model is optimized through the validation set. The multi-source data collected in real time is processed according to the data preprocessing steps and input into the trained LSTM model. The model outputs the predicted equipment failure probability P fault , when P fault When T1 < P, it is judged as a low risk level, indicating that the harmonic condition in the power grid is relatively good; when T1 < P fault When ≤T2, it is judged as medium risk level, and the power grid has a certain degree of harmonic problem; when When , it is judged as a high risk level, indicating that the power grid harmonic pollution is relatively serious, where T1 and T2 are thresholds set based on industry standards.

10. The multi-source fusion smart grid harmonic dynamic management system based on harmonic impedance according to claim 1, characterized in that: The comprehensive assessment and management decision module formulates the harmonic management plan based on the assessment results: Low risk level: Adopt a regular monitoring strategy to continuously track the grid harmonics status and perform preventive maintenance on equipment. Medium risk level: In addition to strengthening frequency monitoring, further analyze harmonic sources to identify the main harmonic-generating equipment or areas, and take measures such as adjusting load distribution, optimizing transformer taps, and installing small harmonic control devices; High-risk level: Immediately initiate emergency control measures, shut down and rectify harmonic source equipment that seriously exceeds the standard, replace large-capacity, high-performance active power filters and static VAR compensators, and install harmonic isolation devices on key equipment that is seriously affected by harmonics.

Images