A maintenance monitoring method for power distribution system operating at low voltage

Abstract

The present invention relates to the technical field of power distribution system maintenance, and discloses a power distribution system maintenance monitoring method for low-voltage operation, the method comprising: performing parameter identification on the power distribution nodes in the power distribution network to be monitored, calculating in real time the deviation between the voltage amplitude prediction value and the actual value of the power distribution nodes, and determining the real-time switch branch state of the power distribution nodes; optimizing and solving the model parameter vector in the fault node monitoring model to obtain the optimized fault node monitoring model, and using the optimized fault node monitoring model to monitor the power distribution nodes in the power distribution network to be monitored. The present invention combines the operating parameters of all distribution lines in the power distribution network, calculates the voltage amplitude prediction value of each distribution node, calculates the real-time switch branch state of the distribution node, and calculates the real-time fault rate of different distribution nodes by using convolution calculation, graph attention weight calculation and fault rate mapping, so as to realize real-time maintenance monitoring of the distribution nodes.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution system maintenance, and in particular to a power distribution system maintenance monitoring method for low-voltage operation. Background Art

[0002] In modern society, power supply plays a vital role in people's life and work. As a bridge connecting the power supply and the terminal equipment, the low-voltage power distribution system undertakes the important task of transmitting electrical energy to various electrical equipment. However, due to factors such as long-term operation and environmental impact of the low-voltage power distribution system, there are various potential failure risks, such as line aging, poor contact, short circuit and other problems. If these failures are not detected and repaired in time, they may cause power outages, safety accidents and even fires and other serious consequences. In response to this problem, the present invention proposes a maintenance monitoring method for a power distribution system for low-voltage operation. Through multi-dimensional data perception and analysis, the problems in the low-voltage power distribution system can be detected and solved in time, the reliability of the power system can be improved, the number of power outages can be reduced, and the power supply quality can be improved. Summary of the invention

[0003] In view of this, the present invention provides a maintenance monitoring method for a power distribution system for low-voltage operation, the purpose of which is to: 1) collect low-voltage power distribution data of different power distribution nodes in the power distribution network to be monitored, and calculate the admittance and impedance parameters of the lines between different power distribution nodes, so as to realize the identification and processing of the operating parameters of the power distribution lines, and calculate the phase angle of the real-time voltage and current of the power distribution nodes based on the identified operating parameters, and calculate the voltage amplitude prediction value of each power distribution node in combination with the operating parameters of all power distribution lines in the power distribution network. If the deviation between the voltage amplitude prediction value and the actual value is too large, it means that the power distribution node is in a non-distributed state, resulting in the inability to predict the voltage amplitude through the power flow calculation method. Realize the real-time switch branch state calculation of distribution nodes in the distribution network; 2) Build a fault node monitoring model, and use convolution calculation, graph attention weight calculation and fault rate mapping to calculate the real-time fault rate of different distribution nodes in turn. In the process of model construction, use linear descent iterative behavior control parameters to control the model parameter vector to have a larger iteration step at the beginning of the iteration, and compare the objective function value of the model parameter vector after the iteration with the objective function value before the iteration, and use the iterative balance operator for optimization feedback. When the iteration obtains a better solution, the convergence speed is accelerated, otherwise the current exploration ability is maintained, and local search is performed to realize the adaptive iteration of the model parameter vector.

[0004] To achieve the above object, the present invention provides a maintenance monitoring method for a power distribution system for low voltage operation, comprising the following steps:

[0005] S1: Collect low-voltage power distribution data, perform parameter identification on the power distribution nodes in the power distribution network to be monitored, and obtain the operating parameters of the power distribution nodes in the power distribution network to be monitored, wherein power flow analysis is the main implementation method of the parameter identification, and the operating parameters include the admittance and impedance of the power distribution nodes;

[0006] S2: Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time to determine the real-time switch branch state of the distribution node;

[0007] S3: Construct a fault node monitoring model, which takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and the real-time fault rate of the distribution node as output, wherein the graph convolutional neural network is the main implementation method of the fault branch monitoring model;

[0008] S4: Optimize and solve the model parameter vector in the fault node monitoring model to obtain an optimized fault node monitoring model, and use the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired.

[0009] As a further improvement method of the present invention:

[0010] Optionally, collecting low-voltage power distribution data in step S1 includes:

[0011] The low-voltage power distribution data of the power distribution nodes in the power distribution network to be monitored are collected, wherein the collection time of the low-voltage power distribution data of each power distribution node is the same, and the representation form of the collected low-voltage power distribution data is:

[0012] {E n =(U n ,P n ,Q n )|n∈[1,N]}

[0013] U n =(U n (1),U n (2),...,U n (m),...,U n (M))

[0014] P n =(P n (1),P n (2),...,P n (m),...,P n (M))

[0015] Q n =(Q n (1),Qn (2),...,Q n (m),...,Q n (M))

[0016] in:

[0017] E n represents the low-voltage power distribution data of the nth power distribution node in the power distribution network to be monitored, and N represents the number of power distribution nodes in the power distribution network to be detected;

[0018] U n represents the voltage data sequence of the nth distribution node in the distribution network to be monitored;

[0019] U n (1),U n (2),...,U n (m),...,U n (M) represents the voltage data of the nth distribution node in the distribution network to be monitored at M consecutive moments, U n (m) represents the voltage data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0020] P n represents the active power data sequence of the nth distribution node in the distribution network to be monitored;

[0021] P n (1),P n (2),...,P n (m),...,P n (M) represents the active power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, P n (m) represents the active power data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0022] Q n Represents the reactive power data sequence of the nth distribution node in the distribution network to be monitored;

[0023] Q n (1),Q n (2),...,Q n (m),...,Q n (M) represents the reactive power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, P n (m) represents the reactive power data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0024] Based on the low-voltage power distribution data, the parameters of the distribution nodes in the distribution network to be monitored are identified to obtain the operating parameters of the distribution nodes in the distribution network to be monitored. In the embodiment of the present invention, the low-voltage power distribution data of each distribution node is collected at the same time.

[0025] Optionally, the performing parameter identification on the distribution nodes in the power distribution network to be monitored to obtain the operating parameters of the distribution nodes in the power distribution network to be monitored includes:

[0026] Parameter identification is performed on the distribution nodes in the distribution network to be monitored to obtain the operating parameters of the distribution nodes in the distribution network to be monitored, wherein power flow analysis is the main implementation method of the parameter identification, and the parameter identification process is as follows:

[0027] S11: Based on the low-voltage power distribution data, the voltage vector U, active power vector P and reactive power vector Q of N distribution nodes in the distribution network to be monitored are constructed:

[0028] U=[U1,U2,...,U n ,...,U N ] T

[0029] P=[P1,P2,...,P n ,...,P N ] T

[0030] Q=[Q1,Q2,...,Q n ,...,Q N ] T

[0031] in:

[0032] T stands for transpose;

[0033] S12: Calculate and obtain the impedance matrix of N distribution nodes in the distribution network to be monitored:

[0034]

[0035] Z=(Z(i,j)) N×N =Re(Z)+Im(Z)

[0036] Re(Z)=(Re[Z(i,j)]) N×N

[0037] Im(Z)=(Im[Z(i,j)]) N×N

[0038] Z(i,j)=Re[Z(i,j)]+Im[Z(i,j)]

[0039] in:

[0040] Z represents the impedance matrix of N distribution nodes in the distribution network to be monitored, Z(i,j) represents the impedance of the line between the i-th distribution node and the j-th distribution node in the distribution network to be monitored, i∈[1,N], j∈[1,N];

[0041] Re[Z(i,j)] represents the real part of the impedance Z(i,j), Im[Z(i,j)] represents the imaginary part of the impedance Z(i,j), Re(Z) represents the real impedance matrix of the N distribution nodes in the distribution network to be monitored, and Im(Z) represents the imaginary impedance matrix of the N distribution nodes in the distribution network to be monitored;

[0042] S13: Calculate and obtain the admittance matrix of N distribution nodes in the distribution network to be monitored:

[0043]

[0044] in:

[0045] B represents the admittance matrix of N distribution nodes in the distribution network to be monitored, B = (B (i, j)) N×N , B(i,j) represents the admittance of the line between the i-th distribution node and the j-th distribution node in the distribution network to be monitored;

[0046] S14: Operation parameters of the distribution nodes in the distribution network to be monitored:

[0047] {S n =(Z n ,B n )|n∈[1,N]}

[0048] in:

[0049] S n represents the operating parameters of the nth distribution node in the distribution network to be monitored;

[0050] Z n represents the nth row in the impedance matrix Z, corresponding to the impedance of the line between the nth distribution node and other distribution nodes;

[0051] B n The nth row in the admittance matrix B corresponds to the admittance of the line between the nth distribution node and other distribution nodes.

[0052] Optionally, in step S2, based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time, including:

[0053] Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time. The deviation calculation process of the nth distribution node in the distribution network to be monitored at the current time t is as follows:

[0054] S21: Obtain the voltage U of the nth distribution node in the distribution network to be monitored at the current time t n (t), active power P n (t) and reactive power Q n (t);

[0055] S22: Obtain the operating parameter S of the nth distribution node in the distribution network to be monitored n , and calculate the phase angle of the line between the nth distribution node and other distribution nodes at the current time t, where the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t is:

[0056] α nj (t)=Im[Z(n,j)]P n (t)-Re[Z(n,j)]Q n (t)

[0057] in:

[0058] α nj (t) represents the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t, j∈[1,N];

[0059] S23: Calculate and obtain the voltage amplitude prediction value u(t) of the nth distribution node in the distribution network to be monitored at the current time t:

[0060] u(t)=u1(t)+u2(t)

[0061]

[0062]

[0063] in:

[0064] u(t) represents the predicted value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t;

[0065] u1(t) represents the real part of the voltage amplitude prediction value u(t), and u2(t) represents the imaginary part of the voltage amplitude prediction value u(t);

[0066] S24: Calculate the deviation c between the predicted value and the actual value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t n (t):

[0067] c n (t) = exp{[u(t)-U n (t)] 2}

[0068] in:

[0069] exp(·) represents an exponential function with a natural constant as the base;

[0070] The real-time switch branch status of the distribution node is determined according to the deviation calculation result.

[0071] Optionally, determining the real-time switch branch state of the power distribution node includes:

[0072] Determine the real-time switch branch status of the distribution node, where the switch branch status of the nth distribution node in the distribution network to be monitored at the current time t is:

[0073]

[0074] in:

[0075] β n (t) represents the switch branch state of the nth distribution node in the distribution network to be monitored at the current time t, β n (t) = 1 means that the nth distribution node in the distribution network to be monitored is in the distribution state at the current time t, β n (t)=0 indicates that the nth distribution node in the distribution network to be monitored is in a non-distribution state at the current time t.

[0076] Optionally, constructing a fault node monitoring model in step S3 includes:

[0077] Construct a fault node monitoring model, which takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and the real-time fault rate of the distribution node as output, wherein the graph convolutional neural network is the main implementation method of the fault branch monitoring model, and the fault branch monitoring model includes an input layer, a convolution calculation layer, a graph attention weight calculation layer, a fault rate mapping layer and an output layer;

[0078] The input layer is used to receive the operating parameters, active power, reactive power and real-time switch branch status of the distribution node;

[0079] The convolution calculation layer is used to perform convolution calculations on the active power, reactive power, and real-time switch branch status of the distribution node;

[0080] The graph attention weight calculation layer is used to extract the attention weight of the operating parameters corresponding to the current distribution node in the distribution network to be monitored;

[0081] The fault rate mapping layer is used to map the convolution calculation results in combination with the attention weights to obtain the real-time fault rate of the distribution node;

[0082] The output layer is used to output the real-time failure rate of the distribution nodes.

[0083] Optionally, in step S4, optimizing and solving the model parameter vector in the fault node monitoring model to obtain an optimized fault node monitoring model includes:

[0084] The model parameter vector in the fault node monitoring model is optimized and solved, and the optimization solution process is as follows:

[0085] S41: Obtain the operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the H group of distribution nodes as training data set data:

[0086] data = {e h =(s h ,p h ,q h ,r h ,y h )|h∈[1,H]}

[0087] in:

[0088] e h represents the hth group of distribution node data in the training data set data;

[0089] s h ,p h ,q h ,r h The operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the hth group of distribution nodes are shown in order. h =1 means that there is a fault in the hth group of distribution nodes, y h =0 means that there is no fault in the hth group of distribution nodes;

[0090] S42: Initialize and generate a model parameter vector θ(0)=(W1(0), b1(0), W2(0)) in the fault node monitoring model, where W1(0), b1(0), W2(0) are respectively the convolution parameter matrix, the bias parameter in the convolution calculation layer, and the mapping parameter matrix in the fault rate mapping layer generated by the initialization;

[0091] S43: Set the current iteration number of the model parameter vector to d, the maximum iteration number to Max, the initial value of d to 0, then the d-th iteration result of the model parameter vector is θ(d)=(W1(d), b1(d), W2(d)); construct the objective function for optimizing the model parameter vector;

[0092] S44: Use the dth iteration result of the model parameter vector as the input of the objective function:

[0093]

[0094] ATT h =[ATT h (1),ATT h (2),...,ATT h (g),...,ATT h (H)]

[0095]

[0096] s h =((z(h,g)) 1×H ,(b(h,g)) 1×H )

[0097] in:

[0098] ATT h represents the attention weight of the operating parameters corresponding to the h-th group of distribution nodes, z(h,g) represents the impedance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, b(h,g) represents the admittance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, g∈[1,H];

[0099] It means that the h-th group of distribution node data is used as the input of the fault node monitoring model constructed based on the model parameter vector θ(d), and the real-time failure rate of the h-th group of distribution nodes output by the fault node monitoring model;

[0100] L(θ(d)) represents the objective function value that takes the dth iteration result of the model parameter vector as input;

[0101] S45: Select the iteration result with the smallest objective function value among the d iteration results as the current optimal search individual

[0102] S46: Iterate the d-th iteration result of the model parameter vector:

[0103]

[0104] in:

[0105] v d represents the iterative behavior control parameter, σ represents the iteration frequency, and σ is set to 2;

[0106] K d represents the iterative balancing operator;

[0107] rand(0,1) represents a random number between 0 and 1, and rand(0,2π) represents a random number between 0 and 2π;

[0108] express The Euclidean distance between

[0109] S47: Let d=d+1, and return to step S44 until the maximum number of iterations is reached;

[0110] S48: Select the iteration result that minimizes the objective function value from the Max iteration results as the solution result of the model parameter vector θ * =(W1,b1,W2), and construct an optimized fault node monitoring model;

[0111] The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, and the distribution nodes that need to be repaired are obtained.

[0112] Optionally, the using of the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired includes:

[0113] The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, where the monitoring process of the nth distribution node in the distribution network to be monitored at the current time t is:

[0114] The input layer receives the operating parameters S of the nth distribution node in the distribution network to be monitored n , Active power P n (t), reactive power Q n (t) and the real-time switch branch state β n (t);

[0115] The convolution calculation layer performs convolution calculation on the active power, reactive power and real-time switch branch status of the nth distribution node in the distribution network to be monitored:

[0116]

[0117] in:

[0118] W1 represents the convolution parameter matrix in the convolution calculation layer;

[0119] * indicates convolution processing;

[0120] b1 represents the bias parameter in the convolution calculation layer;

[0121] represents the convolution calculation result of the nth distribution node in the distribution network to be monitored;

[0122] The attention weight calculation layer extracts the operating parameters S corresponding to the nth distribution node n Attention weights in the distribution network to be monitored:

[0123] att n =[att n (1),att n (2),...,att n (j),...,att n (N)]

[0124]

[0125] in:

[0126] att n Indicates the operating parameter S corresponding to the nth distribution node n Attention weights in the distribution network to be monitored;

[0127] att n (j) represents the weight of the operating parameters of the j-th distribution node in the distribution network to be monitored on the n-th distribution node;

[0128] The failure rate mapping layer combines the attention weights to calculate the convolution results Mapping is performed to obtain the real-time failure rate of the nth distribution node

[0129]

[0130] in:

[0131] represents the real-time failure rate of the nth distribution node in the distribution network to be monitored at the current time t;

[0132] W2 represents the mapping parameter matrix in the failure rate mapping layer;

[0133] The output layer outputs the real-time failure rate of the nth distribution node

[0134] If the real-time failure rate of a distribution node is higher than a preset threshold, the distribution node is marked as a distribution node that needs maintenance and maintenance is performed to achieve real-time maintenance monitoring of the distribution node.

[0135] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0136] A memory storing at least one instruction;

[0137] Communication interface, enabling electronic equipment to communicate; and

[0138] The processor executes the instructions stored in the memory to implement the above-mentioned distribution system maintenance monitoring method for low-voltage operation.

[0139] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned distribution system maintenance monitoring method for low-voltage operation.

[0140] Compared with the prior art, the present invention proposes a maintenance monitoring method for a power distribution system operating at low voltage, which has the following advantages:

[0141] First, this scheme proposes a method for calculating the distribution line parameters and the real-time switch branch status in the distribution network to obtain the operating parameters S of the nth distribution node in the distribution network to be monitored. n , and calculate the phase angle of the line between the nth distribution node and other distribution nodes at the current time t, where the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t is:

[0142] α nj (t)=Im[Z(n,j)]P n (t)-Re[Z(n,j)]Q n (t)

[0143] Where: α nj (t) represents the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t, j∈[1,N]; the voltage amplitude prediction value u(t) of the nth distribution node in the distribution network to be monitored at the current time t is calculated:

[0144] u(t)=u1(t)+u2(t)

[0145]

[0146] Where: u(t) represents the predicted value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t; u1(t) represents the real part of the predicted value of the voltage amplitude u(t), and u2(t) represents the imaginary part of the predicted value of the voltage amplitude u(t); the deviation c between the predicted value and the actual value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t is calculated n (t):

[0147] c n (t) = exp{[u(t)-U n (t)] 2}

[0148] Where: exp(·) represents an exponential function with a natural constant as the base; the real-time switch branch state of the distribution node is determined based on the deviation calculation result. This scheme collects low-voltage distribution data of different distribution nodes in the distribution network to be monitored, and calculates the admittance and impedance parameters of the lines between different distribution nodes to realize the identification and processing of the distribution line operation parameters, and based on the identified operation parameters, calculates the phase angle of the real-time voltage and current of the distribution node, and combines the operation parameters of all distribution lines in the distribution network to calculate the voltage amplitude prediction value of each distribution node. If the deviation between the voltage amplitude prediction value and the actual value is too large, it means that the distribution node is in a non-distributed state, resulting in the inability to predict the voltage amplitude through the flow calculation method, and realizes the real-time switch branch state calculation of the distribution node in the distribution network.

[0149] At the same time, this scheme proposes a model parameter vector optimization solution method to optimize the model parameter vector in the fault node monitoring model. The optimization solution process is: obtain the operating parameters, active power, reactive power, real-time switch branch status and fault detection results of H group distribution nodes as training data set data:

[0150] data = {e h =(s h ,p h ,q h ,r h ,y h )|h∈[1,H]}

[0151] Where: e h represents the hth group of distribution node data in the training data set data; s h ,p h ,q h ,r h The operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the hth group of distribution nodes are shown in order. h =1 means that there is a fault in the hth group of distribution nodes, y h =0 indicates that there is no fault in the h-th group of distribution nodes; initialize and generate the model parameter vector θ(0)=(W1(0),b1(0),W2(0)) in the fault node monitoring model, where W1(0),b1(0),W2(0) are the convolution parameter matrix, bias parameter in the convolution calculation layer and the mapping parameter matrix in the fault rate mapping layer respectively; set the current iteration number of the model parameter vector to d, the maximum iteration number to Max, and the initial value of d to 0, then the d-th iteration result of the model parameter vector is θ(d)=(W1(d),b1(d),W2(d)); construct the objective function for optimizing the model parameter vector; take the d-th iteration result of the model parameter vector as the input of the objective function:

[0152]

[0153] ATT h =[ATT h (1),ATT h (2),...,ATT h (g),...,ATT h (H)]

[0154]

[0155] s h =((z(h,g)) 1×H ,(b(h,g)) 1×H )

[0156] Among them: ATT h represents the attention weight of the operating parameters corresponding to the h-th group of distribution nodes, z(h,g) represents the impedance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, b(h,g) represents the admittance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, g∈[1,H]; = L(θ(d)) represents the objective function value of the dth iteration result of the model parameter vector as input; the iteration result with the smallest objective function value among the dth iteration results is selected as the current optimal search individual. Iterate the d-th iteration result of the model parameter vector:

[0157]

[0158] Where: v d represents the iterative behavior control parameter, σ represents the iterative frequency, and σ is set to 2; K d represents the iterative balance operator; rand(0,1) represents a random number between 0 and 1, and rand(0,2π) represents a random number between 0 and 2π; express The Euclidean distance between them; let d = d + 1 until the maximum number of iterations is reached; select the iteration result that minimizes the objective function value from the Max iteration results as the solution result of the model parameter vector θ *=(W1,b1,E2), and construct an optimized fault node monitoring model; use the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored, and obtain the distribution nodes that need to be repaired. This scheme constructs a fault node monitoring model, and successively uses convolution calculation, graph attention weight calculation and fault rate mapping to calculate the real-time fault rate of different distribution nodes. In the process of model construction, the linear descent iterative behavior control parameter is used to control the model parameter vector to have a larger iteration step at the beginning of the iteration, and the objective function value of the model parameter vector after the iteration is compared with the objective function value before the iteration. The iterative balance operator is used for optimization feedback. When the iteration obtains a better solution, the convergence speed is accelerated, otherwise the current exploration ability is maintained, and local search is performed to realize the adaptive iteration of the model parameter vector. BRIEF DESCRIPTION OF THE DRAWINGS

[0159] Figure 1 A schematic flow chart of a maintenance monitoring method for a power distribution system operating at low voltage provided by one embodiment of the present invention;

[0160] Figure 2 A schematic diagram of the structure of an electronic device for implementing a maintenance monitoring method for a power distribution system operating at low voltage provided by an embodiment of the present invention.

[0161] In the figure: 1 electronic device, 10 processor, 11 memory, 12 program, 13 communication interface.

[0162] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0163] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0164] The embodiment of the present application provides a distribution system maintenance monitoring method for low-voltage operation. The execution subject of the distribution system maintenance monitoring method for low-voltage operation includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the distribution system maintenance monitoring method for low-voltage operation can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0165] Embodiment 1:

[0166] S1: Collect low-voltage power distribution data, perform parameter identification on the power distribution nodes in the power distribution network to be monitored, and obtain the operating parameters of the power distribution nodes in the power distribution network to be monitored.

[0167] The low voltage power distribution data is collected in step S1, including:

[0168] The low-voltage power distribution data of the power distribution nodes in the power distribution network to be monitored are collected, wherein the collection time of the low-voltage power distribution data of each power distribution node is the same, and the representation form of the collected low-voltage power distribution data is:

[0169] {E n =(U n ,P n ,Q n )|n∈[1,N]}

[0170] U n =(U n (1),U n (2),...,U n (m),...,U n (M))

[0171] P n =(P n (1),P n (2),...,P n (m),...,P n (M))

[0172] Q n =(Q n (1),Q n (2),...,Q n (m),...,Q n (M))

[0173] in:

[0174] E n represents the low-voltage power distribution data of the nth power distribution node in the power distribution network to be monitored, and N represents the number of power distribution nodes in the power distribution network to be detected;

[0175] U n represents the voltage data sequence of the nth distribution node in the distribution network to be monitored;

[0176] U n (1),U n (2),...,U n (m),...,U n (M) represents the voltage data of the nth distribution node in the distribution network to be monitored at M consecutive moments, U n (m) represents the voltage data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0177] P nrepresents the active power data sequence of the nth distribution node in the distribution network to be monitored;

[0178] P n (1),P n (2),...,P n (m),...,P n (M) represents the active power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, P n (m) represents the active power data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0179] Q n Represents the reactive power data sequence of the nth distribution node in the distribution network to be monitored;

[0180] Q n (1),Q n (2),...,Q n (m),...,Q n (M) represents the reactive power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, P n (m) represents the reactive power data of the nth distribution node in the distribution network to be monitored at the mth moment;

[0181] Based on the low-voltage power distribution data, parameters of the distribution nodes in the distribution network to be monitored are identified to obtain the operating parameters of the distribution nodes in the distribution network to be monitored.

[0182] The step of performing parameter identification on the distribution nodes in the distribution network to be monitored to obtain the operating parameters of the distribution nodes in the distribution network to be monitored includes:

[0183] Parameter identification is performed on the distribution nodes in the distribution network to be monitored to obtain the operating parameters of the distribution nodes in the distribution network to be monitored, wherein power flow analysis is the main implementation method of the parameter identification, and the parameter identification process is as follows:

[0184] S11: Based on the low-voltage power distribution data, the voltage vector U, active power vector P and reactive power vector Q of N distribution nodes in the distribution network to be monitored are constructed:

[0185] U=[U1,U2,...,U n ,...,U N ] T

[0186] P=[P1,P2,...,P n ,...,P N ] T

[0187] Q=[Q1,Q2,...,Qn ,...,Q N ] T

[0188] in:

[0189] T stands for transpose;

[0190] S12: Calculate and obtain the impedance matrix of N distribution nodes in the distribution network to be monitored:

[0191]

[0192] Z=(Z(i,j)) N×N =Re(Z)+Im(Z)

[0193] Re(Z)=(Re[Z(i,j)]) N×N

[0194] Im(Z)=(Im[Z(i,j)]) N×N

[0195] Z(i,j)=Re[Z(i,j)]+Im[Z(i,j)]

[0196] in:

[0197] Z represents the impedance matrix of N distribution nodes in the distribution network to be monitored, Z(i,j) represents the impedance of the line between the i-th distribution node and the j-th distribution node in the distribution network to be monitored, i∈[1,N], j∈[1,N];

[0198] Re[Z(i,j)] represents the real part of the impedance Z(i,j), Im[Z(i,j)] represents the imaginary part of the impedance Z(i,j), Re(Z) represents the real impedance matrix of the N distribution nodes in the distribution network to be monitored, and Im(Z) represents the imaginary impedance matrix of the N distribution nodes in the distribution network to be monitored;

[0199] S13: Calculate and obtain the admittance matrix of N distribution nodes in the distribution network to be monitored:

[0200]

[0201] in:

[0202] B represents the admittance matrix of N distribution nodes in the distribution network to be monitored, B = (B (i, j)) N×N , B(i,j) represents the admittance of the line between the i-th distribution node and the j-th distribution node in the distribution network to be monitored;

[0203] S14: Operation parameters of the distribution nodes in the distribution network to be monitored:

[0204] {S n =(Z n ,B n )|n∈[1,N]}

[0205] in:

[0206] S n represents the operating parameters of the nth distribution node in the distribution network to be monitored;

[0207] Z n represents the nth row in the impedance matrix Z, corresponding to the impedance of the line between the nth distribution node and other distribution nodes;

[0208] B n The nth row in the admittance matrix B corresponds to the admittance of the line between the nth distribution node and other distribution nodes.

[0209] S2: Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time to determine the real-time switch branch state of the distribution node.

[0210] In step S2, based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time, including:

[0211] Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time. The deviation calculation process of the nth distribution node in the distribution network to be monitored at the current time t is as follows:

[0212] S21: Obtain the voltage U of the nth distribution node in the distribution network to be monitored at the current time t n (t), active power P n (t) and reactive power Q b (t);

[0213] S22: Obtain the operating parameter S of the nth distribution node in the distribution network to be monitored n , and calculate the phase angle of the line between the nth distribution node and other distribution nodes at the current time t, where the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t is:

[0214] α nj (t)=Im[Z(n,j)]P n (t)-Re[Z(n,j)]Q n (t)

[0215] in:

[0216] α nj(t) represents the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t, j∈[1,N];

[0217] S23: Calculate and obtain the voltage amplitude prediction value u(t) of the nth distribution node in the distribution network to be monitored at the current time t:

[0218] u(t)=u1(t)+u2(t)

[0219]

[0220] in:

[0221] u(t) represents the predicted value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t;

[0222] u1(t) represents the real part of the voltage amplitude prediction value u(t), and u2(t) represents the imaginary part of the voltage amplitude prediction value u(t);

[0223] S24: Calculate the deviation c between the predicted value and the actual value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t n (t):

[0224] c n (t) = exp{[u(t)-U n (t)] 2}

[0225] in:

[0226] exp(·) represents an exponential function with a natural constant as the base;

[0227] The real-time switch branch status of the distribution node is determined according to the deviation calculation result.

[0228] Determining the real-time switch branch state of the power distribution node includes:

[0229] Determine the real-time switch branch status of the distribution node, where the switch branch status of the nth distribution node in the distribution network to be monitored at the current time t is:

[0230]

[0231] in:

[0232] β n (t) represents the switch branch state of the nth distribution node in the distribution network to be monitored at the current time t, β n (t) = 1 means that the nth distribution node in the distribution network to be monitored is in the distribution state at the current time t, β n(t)=0 indicates that the nth distribution node in the distribution network to be monitored is in a non-distribution state at the current time t.

[0233] S3: Construct a fault node monitoring model, wherein the fault node monitoring model takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and takes the real-time fault rate of the distribution node as output.

[0234] The fault node monitoring model is constructed in step S3, including:

[0235] Construct a fault node monitoring model, which takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and the real-time fault rate of the distribution node as output, wherein the graph convolutional neural network is the main implementation method of the fault branch monitoring model, and the fault branch monitoring model includes an input layer, a convolution calculation layer, a graph attention weight calculation layer, a fault rate mapping layer and an output layer;

[0236] The input layer is used to receive the operating parameters, active power, reactive power and real-time switch branch status of the distribution node;

[0237] The convolution calculation layer is used to perform convolution calculations on the active power, reactive power, and real-time switch branch status of the distribution node;

[0238] The graph attention weight calculation layer is used to extract the attention weight of the operating parameters corresponding to the current distribution node in the distribution network to be monitored;

[0239] The fault rate mapping layer is used to map the convolution calculation results in combination with the attention weights to obtain the real-time fault rate of the distribution node;

[0240] The output layer is used to output the real-time failure rate of the distribution nodes.

[0241] S4: Optimize and solve the model parameter vector in the fault node monitoring model to obtain an optimized fault node monitoring model, and use the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired.

[0242] In the step S4, the model parameter vector in the fault node monitoring model is optimized and solved to obtain an optimized fault node monitoring model, including:

[0243] The model parameter vector in the fault node monitoring model is optimized and solved, and the optimization solution process is as follows:

[0244] S41: Obtain the operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the H group of distribution nodes as training data set data:

[0245] data = {e h =(s h ,p h ,q h ,r h ,y h )|h∈[1,H]}

[0246] in:

[0247] e h represents the hth group of distribution node data in the training data set data;

[0248] s h ,p h ,q h ,r h The operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the hth group of distribution nodes are shown in order. h =1 means that there is a fault in the hth group of distribution nodes, y h =0 means that there is no fault in the hth group of distribution nodes;

[0249] S42: Initialize and generate a model parameter vector θ(0)=(W1(0), b1(0), W2(0)) in the fault node monitoring model, where W1(0), b1(0), W2(0) are respectively the convolution parameter matrix, the bias parameter in the convolution calculation layer, and the mapping parameter matrix in the fault rate mapping layer generated by the initialization;

[0250] S43: Set the current iteration number of the model parameter vector to d, the maximum iteration number to Max, the initial value of d to 0, then the d-th iteration result of the model parameter vector is θ(d)=(W1(d), b1(d), W2(d)); construct the objective function for optimizing the model parameter vector;

[0251] S44: Use the dth iteration result of the model parameter vector as the input of the objective function:

[0252]

[0253] ATT h =[ATT h (1),ATT h (2),...,ATT h (g),...,ATT h (H)]

[0254]

[0255] s h =((z(h,g)) 1×H ,(b(h,g))1×H )

[0256] in:

[0257] ATT h represents the attention weight of the operating parameters corresponding to the h-th group of distribution nodes, z(h,g) represents the impedance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, b(h,g) represents the admittance of the line between the h-th group of distribution nodes and the g-th group of distribution nodes, g∈[1,H];

[0258] It means that the h-th group of distribution node data is used as the input of the fault node monitoring model constructed based on the model parameter vector θ(d), and the real-time failure rate of the h-th group of distribution nodes output by the fault node monitoring model;

[0259] L(θ(d)) represents the objective function value that takes the dth iteration result of the model parameter vector as input;

[0260] S45: Select the iteration result with the smallest objective function value among the d iteration results as the current optimal search individual

[0261] S46: Iterate the d-th iteration result of the model parameter vector:

[0262]

[0263]

[0264] in:

[0265] v d represents the iterative behavior control parameter, σ represents the iteration frequency, and σ is set to 2;

[0266] K d represents the iterative balancing operator;

[0267] rand(0,1) represents a random number between 0 and 1, and rand(0,2π) represents a random number between 0 and 2π;

[0268] express The Euclidean distance between

[0269] S47: Let d=d+1, and return to step S44 until the maximum number of iterations is reached;

[0270] S48: Select the iteration result that minimizes the objective function value from the Max iteration results as the solution result of the model parameter vector θ * =(W1,b1,W2), and construct an optimized fault node monitoring model;

[0271] The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, and the distribution nodes that need to be repaired are obtained.

[0272] The method of using the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired includes:

[0273] The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, where the monitoring process of the nth distribution node in the distribution network to be monitored at the current time t is:

[0274] The input layer receives the operating parameters S of the nth distribution node in the distribution network to be monitored n , Active power P n (t), reactive power Q n (t) and the real-time switch branch state β n (t);

[0275] The convolution calculation layer performs convolution calculation on the active power, reactive power and real-time switch branch status of the nth distribution node in the distribution network to be monitored:

[0276]

[0277] in:

[0278] W1 represents the convolution parameter matrix in the convolution calculation layer;

[0279] * indicates convolution processing;

[0280] b1 represents the bias parameter in the convolution calculation layer;

[0281] represents the convolution calculation result of the nth distribution node in the distribution network to be monitored;

[0282] The attention weight calculation layer extracts the operating parameters S corresponding to the nth distribution node n Attention weights in the distribution network to be monitored:

[0283] att n =[att n (1),att n (2),...,att n (j),...,att n (N)]

[0284]

[0285] in:

[0286] att n Indicates the operating parameter S corresponding to the nth distribution node n Attention weights in the distribution network to be monitored;

[0287] att n (j) represents the weight of the operating parameters of the j-th distribution node in the distribution network to be monitored on the n-th distribution node;

[0288] The failure rate mapping layer combines the attention weights to calculate the convolution results Mapping is performed to obtain the real-time failure rate of the nth distribution node

[0289]

[0290] in:

[0291] represents the real-time failure rate of the nth distribution node in the distribution network to be monitored at the current time t;

[0292] W2 represents the mapping parameter matrix in the failure rate mapping layer;

[0293] The output layer outputs the real-time failure rate of the nth distribution node

[0294] If the real-time failure rate of a distribution node is higher than a preset threshold, the distribution node is marked as a distribution node that needs maintenance and maintenance is performed to achieve real-time maintenance monitoring of the distribution node.

[0295] Embodiment 2:

[0296] like Figure 2 , which is a schematic diagram of the structure of an electronic device for implementing a maintenance monitoring method for a power distribution system operating at low voltage provided by an embodiment of the present invention.

[0297] The electronic device 1 may include a processor 10 , a memory 11 , a communication interface 13 and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10 , such as a program 12 .

[0298] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a mobile hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 may be used not only to store application software and various types of data installed in the electronic device 1, such as the code of the program 12, etc., but also to temporarily store data that has been output or is to be output.

[0299] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as a program 12 for implementing maintenance monitoring of a power distribution system for low-voltage operation), and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0300] The communication interface 13 may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices, and to achieve connection and communication between internal components of the electronic device.

[0301] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0302] Figure 2 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 2 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0303] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0304] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0305] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0306] The program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0307] Collect low-voltage power distribution data, perform parameter identification on the power distribution nodes in the power distribution network to be monitored, and obtain the operating parameters of the power distribution nodes in the power distribution network to be monitored;

[0308] Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time to determine the real-time switch branch state of the distribution node;

[0309] Build a fault node monitoring model;

[0310] The model parameter vector in the fault node monitoring model is optimized and solved to obtain the optimized fault node monitoring model. The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired.

[0311] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figure 1 to Figure 2 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0312] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0313] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0314] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A maintenance monitoring method for a power distribution system operating at low voltage, characterized in that: The method comprises: S1: Collect low-voltage power distribution data, identify parameters of the power distribution nodes in the power distribution network to be monitored, and obtain the operating parameters of the power distribution nodes in the power distribution network to be monitored; S2: Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time to determine the real-time switch branch state of the distribution node; S3: construct a fault node monitoring model, wherein the fault node monitoring model takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and takes the real-time fault rate of the distribution node as output; S4: Optimizing and solving the model parameter vector in the fault node monitoring model to obtain an optimized fault node monitoring model, and using the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired; The fault node monitoring model is constructed in step S3, including: Construct a fault node monitoring model, which takes the operating parameters, active power, reactive power and real-time switch branch status of the distribution node as input and the real-time fault rate of the distribution node as output. The fault branch monitoring model includes an input layer, a convolution calculation layer, a graph attention weight calculation layer, a fault rate mapping layer and an output layer; The input layer is used to receive the operating parameters, active power, reactive power and real-time switch branch status of the distribution node; The convolution calculation layer is used to perform convolution calculations on the active power, reactive power, and real-time switch branch status of the distribution node; The graph attention weight calculation layer is used to extract the attention weight of the operating parameters corresponding to the current distribution node in the distribution network to be monitored; The fault rate mapping layer is used to map the convolution calculation results in combination with the attention weights to obtain the real-time fault rate of the distribution node; The output layer is used to output the real-time failure rate of the power distribution nodes; In the step S4, the model parameter vector in the fault node monitoring model is optimized and solved to obtain an optimized fault node monitoring model, including: The model parameter vector in the fault node monitoring model is optimized and solved, and the optimization solution process is as follows: S41: Obtain the operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the H group of distribution nodes as training data set data: in: represents the hth group of distribution node data in the training data set data; The operating parameters, active power, reactive power, real-time switch branch status and fault detection results of the h-th group of distribution nodes are shown in order. Indicates that there is a fault in the hth group of distribution nodes. Indicates that there is no fault in the hth group of distribution nodes; S42: Initialize and generate the model parameter vector in the fault node monitoring model ,in The convolution parameter matrix, the bias parameter in the convolution calculation layer and the mapping parameter matrix in the failure rate mapping layer are initialized and generated in turn; S43: Set the current iteration number of the model parameter vector to d, the maximum iteration number to Max, and the initial value of d to 0. Then the dth iteration result of the model parameter vector is ; Construct the objective function for optimizing the model parameter vector; S44: Use the dth iteration result of the model parameter vector as the input of the objective function: in: represents the attention weight of the operating parameters corresponding to the h-th group of distribution nodes, Represents the hth group of distribution nodes and the The impedance of the line between the group distribution nodes, Represents the hth group of distribution nodes and the The admittance of the lines between the group distribution nodes, ; Indicates that the hth group of distribution node data is used as the model parameter vector The input of the constructed fault node monitoring model, the real-time failure rate of the hth group of distribution nodes output by the fault node monitoring model; It represents the objective function value that takes the dth iteration result of the model parameter vector as input; S45: Select the iteration result with the smallest objective function value among the d iteration results as the current optimal search individual ; S46: Iterate the d-th iteration result of the model parameter vector: in: represents the iterative behavior control parameter, represents the iteration frequency, Set to 2; represents the iterative balancing operator; Represents a random number between 0 and 1. Indicates 0- A random number between express The Euclidean distance between S47: Set d=d+1, and return to step S44 until the maximum number of iterations is reached; S48: Select the iteration result that minimizes the objective function value from the Max iteration results as the solution result of the model parameter vector , and build an optimized fault node monitoring model; The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, and the distribution nodes that need to be repaired are obtained; The method of using the optimized fault node monitoring model to monitor the distribution nodes in the distribution network to be monitored to obtain the distribution nodes that need to be repaired includes: The optimized fault node monitoring model is used to monitor the distribution nodes in the distribution network to be monitored, where the monitoring process of the nth distribution node in the distribution network to be monitored at the current time t is: The input layer receives the operating parameters of the nth distribution node in the distribution network to be monitored. , Active Power , reactive power And real-time switch branch status ; The convolution calculation layer performs convolution calculation on the active power, reactive power and real-time switch branch status of the nth distribution node in the distribution network to be monitored: in: Represents the convolution parameter matrix in the convolution calculation layer; * indicates convolution processing; Represents the bias parameters in the convolution calculation layer; represents the convolution calculation result of the nth distribution node in the distribution network to be monitored; The attention weight calculation layer extracts the operating parameters corresponding to the nth distribution node Attention weights in the distribution network to be monitored: in: Indicates the operating parameters corresponding to the nth distribution node Attention weights in the distribution network to be monitored; represents the weight of the operating parameters of the jth distribution node in the distribution network to be monitored that affects the nth distribution node; The failure rate mapping layer combines the attention weights to calculate the convolution results Mapping is performed to obtain the real-time failure rate of the nth distribution node : in: represents the real-time failure rate of the nth distribution node in the distribution network to be monitored at the current time t; represents the mapping parameter matrix in the failure rate mapping layer; The output layer outputs the real-time failure rate of the nth distribution node ; If the real-time failure rate of a distribution node is higher than a preset threshold, the distribution node is marked as a distribution node that needs maintenance and maintenance is performed.

2. A maintenance monitoring method for a power distribution system for low voltage operation as claimed in claim 1, characterized in that: The low voltage power distribution data is collected in step S1, including: The low-voltage power distribution data of the power distribution nodes in the power distribution network to be monitored are collected, wherein the collection time of the low-voltage power distribution data of each power distribution node is the same, and the representation form of the collected low-voltage power distribution data is: in: represents the low-voltage power distribution data of the nth power distribution node in the power distribution network to be monitored, and N represents the number of power distribution nodes in the power distribution network to be detected; represents the voltage data sequence of the nth distribution node in the distribution network to be monitored; represents the voltage data of the nth distribution node in the distribution network to be monitored at M consecutive moments, Represents the voltage data of the nth distribution node in the distribution network to be monitored at the mth moment; represents the active power data sequence of the nth distribution node in the distribution network to be monitored; Represents the active power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, Represents the active power data of the nth distribution node in the distribution network to be monitored at the mth moment; Represents the reactive power data sequence of the nth distribution node in the distribution network to be monitored; Represents the reactive power data of the nth distribution node in the distribution network to be monitored at M consecutive moments, Represents the reactive power data of the nth distribution node in the distribution network to be monitored at the mth moment; Based on the low-voltage power distribution data, parameters of the distribution nodes in the distribution network to be monitored are identified to obtain the operating parameters of the distribution nodes in the distribution network to be monitored.

3. A maintenance monitoring method for a power distribution system for low voltage operation as claimed in claim 2, characterized in that: The step of performing parameter identification on the distribution nodes in the distribution network to be monitored to obtain the operating parameters of the distribution nodes in the distribution network to be monitored includes: Parameter identification is performed on the distribution nodes in the distribution network to be monitored to obtain the operating parameters of the distribution nodes in the distribution network to be monitored. The parameter identification process is as follows: S11: Based on the low-voltage power distribution data, the voltage vector U, the active power vector P and the reactive power vector Q of the N distribution nodes in the distribution network to be monitored are constructed: in: T stands for transpose; S12: Calculate and obtain the impedance matrix of N distribution nodes in the distribution network to be monitored: in: represents the impedance matrix of N distribution nodes in the distribution network to be monitored, represents the impedance of the line between the ith distribution node and the jth distribution node in the distribution network to be monitored, , ; Impedance The real part of Impedance The imaginary part of represents the real impedance matrix of N distribution nodes in the distribution network to be monitored, represents the imaginary impedance matrix of N distribution nodes in the distribution network to be monitored; S13: Calculate and obtain the admittance matrix of N distribution nodes in the distribution network to be monitored: in: B represents the admittance matrix of N distribution nodes in the distribution network to be monitored, , represents the admittance of the line between the ith distribution node and the jth distribution node in the distribution network to be monitored; S14: Operation parameters of the distribution nodes in the distribution network to be monitored: in: represents the operating parameters of the nth distribution node in the distribution network to be monitored; Impedance matrix The nth row in corresponds to the impedance of the line between the nth distribution node and other distribution nodes; It represents the nth row in the admittance matrix B, corresponding to the admittance of the line between the nth distribution node and other distribution nodes.

4. A maintenance monitoring method for a power distribution system for low voltage operation as claimed in claim 3, characterized in that: In step S2, based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time, including: Based on the identified operating parameters, the deviation between the predicted value and the actual value of the voltage amplitude of the distribution node is calculated in real time. The deviation calculation process of the nth distribution node in the distribution network to be monitored at the current time t is as follows: S21: Obtain the voltage of the nth distribution node in the distribution network to be monitored at the current time t , Active Power And reactive power ; S22: Obtain the operating parameters of the nth distribution node in the distribution network to be monitored , and calculate the phase angle of the line between the nth distribution node and other distribution nodes at the current time t, where the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t is: in: represents the phase angle of the line between the nth distribution node and the jth distribution node in the distribution network to be monitored at the current time t, ; S23: Calculate the predicted value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t : in: represents the predicted value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t; Indicates the predicted value of voltage amplitude The real part of Indicates the predicted value of voltage amplitude The imaginary part of S24: Calculate the deviation between the predicted value and the actual value of the voltage amplitude of the nth distribution node in the distribution network to be monitored at the current time t : in: represents an exponential function with a natural constant as base; The real-time switch branch status of the distribution node is determined according to the deviation calculation result.

5. A maintenance monitoring method for a power distribution system operating at low voltage as claimed in claim 4, characterized in that: Determining the real-time switch branch state of the power distribution node includes: Determine the real-time switch branch status of the distribution node, where the switch branch status of the nth distribution node in the distribution network to be monitored at the current time t is: in: represents the switch branch state of the nth distribution node in the distribution network to be monitored at the current time t, It means that the nth distribution node in the distribution network to be monitored is in the distribution state at the current time t. It indicates that the nth distribution node in the distribution network to be monitored is in a non-distributed state at the current time t.

Images