Power distribution network self-healing method and system based on distribution network scada

Abstract

The invention discloses a distribution network scada-based power distribution network self-healing method and system, and the method comprises the steps: carrying out the data collection according to a scada system, and carrying out the preprocessing; and carrying out feature extraction on the preprocessed data, and detecting an abnormal condition by using an optimized LOF algorithm. And fault isolation is carried out, a power grid topological structure is reconstructed, and self-healing of the power distribution network is realized. According to the distribution network scada-based power distribution network self-healing method and system provided by the invention, feature extraction is performed on the data, and the optimized LOF algorithm is introduced to be combined with the genetic algorithm to perform anomaly detection, so that the accuracy and efficiency of fault detection are remarkably improved, and a fault point can be quickly identified. Through an SCADA system and an automatic switch, automatic isolation of faults is realized, linear programming and an ant colony algorithm are combined, a power grid reconstruction scheme is optimized, load balance is ensured, and reliability and stability of a power grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power automation, and in particular to a distribution network self-healing method and system based on distribution network SCADA. Background Art

[0002] With the development of smart grid technology, the automation and intelligence of distribution networks are constantly improving, and SCADA (Supervisory Control and Data Acquisition System) has been widely used in distribution networks. The SCADA system monitors and controls the operating status of the power grid in real time through sensors and communication networks, providing a large amount of operating data, laying the foundation for the intelligent management and maintenance of the power grid. In the prior art, the self-healing method of the distribution network mainly relies on manual operation and pre-set rules, and achieves fault isolation and grid reconstruction through the operation of switches and circuit breakers. However, traditional self-healing methods have problems such as slow response speed and low intelligence, which are difficult to meet the needs of modern power grids for fast and accurate fault handling.

[0003] At present, the research and application of distribution network self-healing technology has made certain progress, mainly including automated switch operation, fault detection and location, load transfer and other aspects. These technologies improve the efficiency and accuracy of fault handling through automated equipment and intelligent algorithms. However, there are still some shortcomings in the existing technology, such as the accuracy and efficiency of the fault detection algorithm needs to be improved, and the optimization method of power grid reconstruction is relatively simple, which is difficult to cope with complex power grid topology and variable load conditions. At the same time, the existing technology lacks flexibility in data processing and analysis, and it is difficult to make full use of the rich data provided by the SCADA system for intelligent decision-making. Therefore, it is of great application value and practical significance to develop a distribution network self-healing method based on the SCADA system, and use the optimization algorithm to improve the efficiency and accuracy of fault detection and power grid reconstruction. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the fault detection algorithms in the prior art usually have the problem of insufficient accuracy and efficiency. The traditional fault isolation and grid reconstruction methods rely on pre-set rules and manual operations, have slow response speeds, and are difficult to cope with complex grid topologies.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a distribution network self-healing method based on distribution network SCADA, comprising: collecting data and performing preprocessing according to the SCADA system.

[0007] Feature extraction is performed on the preprocessed data, and anomalies are detected using the optimized LOF algorithm.

[0008] Conduct fault isolation, reconstruct the power grid topology, and achieve self-healing of the distribution network.

[0009] As a preferred solution of the distribution network self-healing method based on distribution network SCADA described in the present invention, the data collection and preprocessing according to the SCADA system include that the SCADA system collects the operation data of the distribution network in real time through sensors and instruments, and transmits the data through a communication network. The collected data includes voltage, current, frequency, power, phase angle and power quality.

[0010] Data preprocessing includes data normalization, which is expressed as:

[0011]

[0012] Among them, x i ′ represents the normalized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

[0013] As a preferred solution of the distribution network self-healing method based on distribution network SCADA described in the present invention, wherein: the feature extraction of the pre-processed data includes, the voltage feature extraction is expressed as:

[0014]

[0015] Among them, T V Represents voltage characteristics, V max Indicates the maximum voltage, V min Indicates the minimum voltage, V arg Indicates the average voltage.

[0016] The current feature extraction is expressed as:

[0017]

[0018] Among them, T I Represents the current characteristics, I rms Represents the RMS value of current, I peak Indicates the peak value of the current.

[0019] The frequency feature extraction formula is expressed as:

[0020] T f =f std

[0021] Among them, T f Represents the frequency characteristics, f std Represents the standard deviation of frequencies.

[0022] The power feature extraction is expressed as:

[0023]

[0024] Among them, T P Represents the active power characteristic, P max Indicates the maximum active power, P min Indicates that there is a minimum value for the public highway, P arg Indicates the average value of active power. Q Represents reactive power characteristics, Q max Indicates the maximum reactive power, Q min Indicates the minimum reactive power, Q arg Indicates the average value of reactive power.

[0025] Phase angle feature extraction is expressed as:

[0026] T θ =cos(θ avg )

[0027] Among them, T θ Represents the phase angle characteristic, θ avg Represents the average phase angle.

[0028] The power quality feature extraction is expressed as:

[0029]

[0030] Among them, T PQ Indicates the power quality characteristics, H total Indicates total harmonic distortion, V rms Indicates the RMS value of voltage.

[0031] As a preferred solution of the distribution network self-healing method based on distribution network SCADA described in the present invention, wherein: the use of the optimized LOF algorithm to detect abnormal conditions includes fusing features, which is expressed as:

[0032]

[0033] Among them, F i represents the normalized features, T i represents the value of feature i, μ i represents the mean of feature i, σ i represents the standard deviation of feature i.

[0034] The calculation formula of LOF algorithm is expressed as:

[0035]

[0036] Among them, lrd k(p) represents the local reachable density, expressed as:

[0037]

[0038] Among them, p represents the point to be detected, N k (p) represents the k nearest neighbor set of point p, reach_dist k (p,o) represents the reachable distance, reach_dist k (p,o)=max{dist(p,o),min_dist k (o)}. dist k (p,o) represents the distance between points p and o, min_dist k (o) represents the kth nearest neighbor distance from point o to its k nearest neighbors.

[0039] Genetic algorithms are used to optimize the calculation process of the LOF algorithm and improve the accuracy and efficiency of the algorithm. The genetic algorithm includes initialization, fitness calculation, selecting some individuals as parents of the next generation according to fitness, performing crossover operations on the selected individuals to generate new individuals, introducing randomness through mutation, and iteration.

[0040] Initialization involves randomly generating an initial population, where each individual represents a locally reachable density calculation scheme. The initial population is {C1,C2,...,C N}, each individual Ci contains a parameter set {α i ,β i The fitness calculation includes a formula based on the local anomaly factor to calculate the local reachable density of each individual in the current population and compare it with other individuals. The fitness function is expressed as:

[0041]

[0042] Among them, f(C i ) represents the fitness function, Indicates the LOF value calculated based on the Ci parameter.

[0043] The optimization parameters are obtained by genetic algorithm, α represents the voltage difference adjustment factor determined by genetic algorithm, and β represents the current difference adjustment factor determined by genetic algorithm. The distance calculation formula after optimization of the optimization parameters is expressed as:

[0044]

[0045] Among them, dist opt (p,o) represents the physical distance between points p and o, V p Represents the voltage at point p, V o Indicates the voltage at point o, I prepresents the current at point p, I o represents the current at point o.

[0046] The local reachable density formula after optimization of the optimization parameters is expressed as:

[0047]

[0048] The optimized local reachable density is used to calculate the LOF value of each point and determine whether it is an abnormal point, which is expressed as:

[0049]

[0050] in, represents the optimized reachable distance,

[0051] As a preferred solution of the distribution network self-healing method based on distribution network scada described in the present invention, the use of the optimized LOF algorithm to detect abnormal conditions also includes combining the fused features and the optimized LOF algorithm to comprehensively calculate the abnormality score, which is expressed as:

[0052]

[0053] Among them, A p represents the comprehensive anomaly score of point p, N represents the number of features, and F i Represents the normalized value of feature i, LOF k (p) represents the anomaly score of point p, λ i represents the weight of feature i.

[0054] Comprehensive abnormality score A p When it is greater than 0.5, the point is considered abnormal, otherwise it is normal.

[0055] As a preferred solution of the distribution network self-healing method based on distribution network SCADA described in the present invention, the fault isolation includes: p >0.5, the fault is isolated, and the SCADA system sends instructions to the automation switch, which disconnects the faulty equipment and lines and cuts off the power supply to the faulty area.

[0056] As a preferred solution of the distribution network self-healing method based on distribution network SCADA described in the present invention, wherein: the reconstructed power grid topology structure includes, the goal is to minimize load imbalance and reconstruction cost, and the power grid reconstruction objective function and constraint conditions based on linear programming are expressed as:

[0057]

[0058] subject to

[0059]

[0060]

[0061]

[0062] in, represents the load of the i-th line after reconstruction, represents the load of the ith line before reconstruction, C j represents the cost of the j-th reconstruction operation, x j A binary variable representing the jth reconstruction operation, 1 indicates that the reconstruction operation is performed, and 0 indicates that the reconstruction operation is not performed. i represents the capacity of the ith line, represents the load of the kth transformer after reconstruction, Capacity k represents the capacity of the kth transformer.

[0063] The ant colony algorithm is introduced to optimize the linear programming-based power grid reconstruction method, initialize the ant colony, and set the initial pheromone concentration. In the power grid topology, each line and transformer is assigned an initial pheromone concentration τ0.

[0064] In the grid reconstruction problem, the choice of each line and transformer constitutes the solution space, and each ant selects a path based on the pheromone concentration τ and heuristic information η. The selection probability P ij It is expressed as:

[0065]

[0066] Among them, τ ij represents the pheromone concentration of line i to j, η ij It represents the heuristic information of line i to j, including line capacity and current load. Allowed represents the set of optional paths for ants, α' represents the pheromone weight, and β' represents the heuristic information weight.

[0067] After each iteration, the pheromone concentration is updated according to the path selection of the ants, expressed as:

[0068]

[0069] Where ρ represents the pheromone volatility coefficient, It represents the amount of pheromone released by the kth ant on the path from i to j.

[0070] The path selection and pheromone updating process is repeated until a high-quality solution is obtained, and the optimal reconstruction plan is determined, including the switches and circuit breakers that need to be operated, as well as the load transfer path.

[0071] A distribution network self-healing system based on distribution network SCADA, characterized by: including:

[0072] The data collection module collects data and performs preprocessing according to the SCADA system.

[0073] The anomaly detection module extracts features from the preprocessed data and uses the optimized LOF algorithm to detect anomalies.

[0074] The self-healing module isolates faults, reconstructs the grid topology, and realizes self-healing of the distribution network.

[0075] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0076] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0077] Beneficial effects of the present invention: The present invention significantly improves the accuracy and efficiency of fault detection by extracting features from data and introducing an optimized LOF algorithm combined with a genetic algorithm for anomaly detection, and can quickly identify the fault point. Through the SCADA system and automated switches, automatic fault isolation is achieved, and combined with linear programming and ant colony algorithms, the power grid reconstruction plan is optimized to ensure load balancing and improve the reliability and stability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0079] Figure 1 An overall flow chart of a distribution network self-healing method and system based on distribution network SCADA is provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0080] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0081] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a distribution network self-healing method based on distribution network SCADA, including:

[0082] S1: Collect data and perform preprocessing according to the SCADA system.

[0083] The SCADA system collects the operating data of the distribution network in real time through sensors and instruments and transmits the data through the communication network. The collected data includes voltage, current, frequency, power, phase angle and power quality.

[0084] Data preprocessing includes data normalization, which is expressed as:

[0085]

[0086] Among them, x i ′ represents the normalized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

[0087] S2: Extract features from the preprocessed data and use the optimized LOF algorithm to detect anomalies.

[0088] Voltage feature extraction is expressed as:

[0089]

[0090] Among them, T V Represents voltage characteristics, V max Indicates the maximum voltage, V min Indicates the minimum voltage, V arg Indicates the average voltage.

[0091] The current feature extraction is expressed as:

[0092]

[0093] Among them, T I Represents the current characteristics, I rms Represents the RMS value of current, I peak Indicates the peak value of the current.

[0094] The frequency feature extraction formula is expressed as:

[0095] T f =f std

[0096] Among them, T f Represents the frequency characteristics, f stdRepresents the standard deviation of frequencies.

[0097] The power feature extraction is expressed as:

[0098]

[0099] Among them, T P Represents the active power characteristic, P max Indicates the maximum active power, P min Indicates that there is a minimum value for the public highway, P arg Indicates the average value of active power. Q Represents reactive power characteristics, Q max Indicates the maximum reactive power, Q min Indicates the minimum reactive power, Q arg Indicates the average value of reactive power.

[0100] The phase angle feature extraction is expressed as:

[0101] T θ =cos(θ avg )

[0102] Among them, T θ Represents the phase angle characteristic, θ avg Represents the average phase angle.

[0103] The power quality feature extraction is expressed as:

[0104]

[0105] Among them, T PQ Indicates the power quality characteristics, H total Indicates total harmonic distortion, V rms Indicates the RMS value of voltage.

[0106] The features are fused and expressed as:

[0107]

[0108] Among them, F i represents the normalized features, T i represents the value of feature i, μ i represents the mean of feature i, σ i represents the standard deviation of feature i.

[0109] The calculation formula of LOF algorithm is expressed as:

[0110]

[0111] Among them, lrd k (p) represents the local reachable density, expressed as:

[0112]

[0113] Among them, p represents the point to be detected, N k (p) represents the k nearest neighbor set of point p, reach_dist k (p,o) represents the reachable distance, reach_dist k (p,o)=max{dist(p,o),min_dist k (o)}. dist k (p,o) represents the distance between points p and o, min_dist k (o) represents the kth nearest neighbor distance from point o to its k nearest neighbors.

[0114] Genetic algorithms are used to optimize the calculation process of the LOF algorithm and improve the accuracy and efficiency of the algorithm. The genetic algorithm includes initialization, fitness calculation, selecting some individuals as parents of the next generation according to fitness, performing crossover operations on the selected individuals to generate new individuals, introducing randomness through mutation, and iteration.

[0115] Initialization involves randomly generating an initial population, where each individual represents a locally reachable density calculation scheme. The initial population is {C1,C2,...,C N}, each individual Ci contains a parameter set {α i ,β i The fitness calculation includes a formula based on the local anomaly factor to calculate the local reachable density of each individual in the current population and compare it with other individuals. The fitness function is expressed as:

[0116]

[0117] Among them, f(C i ) represents the fitness function, Indicates the LOF value calculated based on the Ci parameter.

[0118] It should be noted that the fitness function is based on the formula of the local anomaly factor, which calculates the local reachable density of each individual in the current population and compares it with other individuals.

[0119] Some individuals are selected as parents of the next generation according to their fitness. The selection strategy is roulette wheel selection to ensure that individuals with high fitness have a greater probability of being selected.

[0120] Perform a crossover operation on the selected individuals to generate new individuals. The crossover operation is a multi-point crossover, and the generated new individuals will inherit some of the characteristics of the parent individuals.

[0121] Perform mutation operations on some new individuals to introduce randomness. The mutation operation can be a numerical change to increase the diversity of the population.

[0122] Furthermore, the genetic algorithm steps are repeated until the predetermined number of iterations is 100. After each iteration, the individuals in the population are updated so that the calculation scheme of the local reachable density is gradually optimized. The optimization parameters are obtained by the genetic algorithm, α represents the voltage difference adjustment factor determined by the genetic algorithm, and β represents the current difference adjustment factor determined by the genetic algorithm. The distance calculation formula after optimization of the optimization parameters is expressed as:

[0123]

[0124] Among them, dist opt (p,o) represents the physical distance between points p and o, V p Represents the voltage at point p, V o Indicates the voltage at point o, I p represents the current at point p, I o represents the current at point o.

[0125] The local reachable density formula after optimization of the optimization parameters is expressed as:

[0126]

[0127] The optimized local reachable density is used to calculate the LOF value of each point and determine whether it is an abnormal point, which is expressed as:

[0128]

[0129] in, represents the optimized reachable distance,

[0130] Combining the fused features and the optimized LOF algorithm, the anomaly score is calculated comprehensively, expressed as:

[0131]

[0132] Among them, A p represents the comprehensive anomaly score of point p, N represents the number of features, and F i Represents the normalized value of feature i, LOF k (p) represents the anomaly score of point p, λ i represents the weight of feature i.

[0133] Comprehensive abnormality score A p When it is greater than 0.5, the point is considered abnormal, otherwise it is normal.

[0134] It should be noted that the genetic algorithm (GA) is an optimization algorithm based on natural selection and genetic mechanisms, which searches for the optimal solution by simulating the biological evolution process. When optimizing the LOF (Local Outlier Factor) algorithm, we first randomly generate the initial population, and each individual represents a possible local reachable density (LRD) calculation scheme. Each individual contains a set of parameters (such as voltage difference adjustment factor and current difference adjustment factor), which will affect the distance calculation of the LOF algorithm. The fitness function is used to evaluate the quality of each individual, and the fitness function is based on the accuracy of the LOF value. Individuals with high fitness are selected as the parents of the next generation, and crossover and mutation operations are performed to generate new individuals. The population is continuously updated iteratively to gradually optimize the calculation formula of the local reachable density.

[0135] Furthermore, the genetic algorithm gradually optimizes parameters through the evolutionary mechanism, which can more accurately adjust the distance calculation in the LOF algorithm and improve the accuracy of anomaly detection. The genetic algorithm has global search capabilities and can avoid falling into the local optimal solution, thereby finding a better combination of parameters and improving the performance of the LOF algorithm. The genetic algorithm does not rely on specific initial parameter settings, and automatically optimizes parameters through the evolutionary process, reducing human intervention and improving the adaptability and robustness of the algorithm.

[0136] S3: Perform fault isolation, reconstruct the grid topology, and achieve self-healing of the distribution network.

[0137] The comprehensive abnormality score A p >0.5, the fault is isolated, and the SCADA system sends instructions to the automation switch, which disconnects the faulty equipment and lines and cuts off the power supply to the faulty area.

[0138] The goal is to minimize load imbalance and reconstruction cost. The objective function and constraints of power grid reconstruction based on linear programming are expressed as:

[0139]

[0140] subject to

[0141]

[0142]

[0143]

[0144] in, represents the load of the i-th line after reconstruction, represents the load of the ith line before reconstruction, C j represents the cost of the j-th reconstruction operation, xj A binary variable representing the jth reconstruction operation, 1 indicates that the reconstruction operation is performed, and 0 indicates that the reconstruction operation is not performed. i represents the capacity of the ith line, represents the load of the kth transformer after reconstruction, Capacity k represents the capacity of the kth transformer.

[0145] The ant colony algorithm is introduced to optimize the linear programming-based power grid reconstruction method, initialize the ant colony, and set the initial pheromone concentration. In the power grid topology, each line and transformer is assigned an initial pheromone concentration τ0.

[0146] In the grid reconstruction problem, the choice of each line and transformer constitutes the solution space, and each ant selects a path based on the pheromone concentration τ and heuristic information η. The selection probability P ij It is expressed as:

[0147]

[0148] Among them, τ ij represents the pheromone concentration of line i to j, η ij It represents the heuristic information of line i to j, including line capacity and current load. Allowed represents the set of optional paths for ants, α' represents the pheromone weight, and β' represents the heuristic information weight.

[0149] After each iteration, the pheromone concentration is updated according to the path selection of the ants, expressed as:

[0150]

[0151] Where ρ represents the pheromone volatility coefficient, It represents the amount of pheromone released by the kth ant on the path from i to j.

[0152] The path selection and pheromone updating process is repeated until a high-quality solution is obtained, and the optimal reconstruction plan is determined, including the switches and circuit breakers that need to be operated, as well as the load transfer path.

[0153] It should be noted that the Ant Colony Optimization (ACO) is an optimization algorithm that simulates the foraging behavior of ants and finds the optimal solution through pheromone updates and path selection. In the power grid reconstruction problem, we combine the ACO with linear programming. First, a linear programming model is established with the goal of minimizing load imbalance and reconstruction cost. Then, the ant colony is initialized and the initial pheromone concentration is assigned on the power grid topology. Each ant selects a path based on the pheromone concentration and heuristic information and calculates the reconstruction plan. Through multiple iterations, the ants continuously update the pheromone concentration on the path and gradually optimize the power grid reconstruction plan. Finally, based on the optimization results of the ACO, the optimal reconstruction plan is determined, including the switches that need to be operated and the load transfer path.

[0154] The ant colony algorithm has global optimization capabilities through the propagation and update mechanism of pheromones, which can effectively avoid falling into local optimal solutions and improve the overall quality of power grid reconstruction solutions. The ant colony algorithm can dynamically adjust the pheromone concentration according to the operating status of the power grid, adapt to different load conditions and topological structures, and ensure the flexibility and robustness of the reconstruction solution. The ant colony algorithm, combined with heuristic information (line capacity and current load), can more intelligently evaluate the pros and cons of path selection and improve the efficiency and accuracy of the reconstruction solution.

[0155] By introducing the ant colony algorithm, the grid reconstruction method can more efficiently handle complex grid topology and load distribution, achieve fast and intelligent grid reconstruction, ensure stable and reliable power supply in non-fault areas, and improve the self-healing capability of the distribution network.

[0156] Furthermore, all weight parameters, coefficients and other imaginary parameters of the present invention are obtained through experiments and set according to the experience of relevant personnel.

[0157] The above embodiments also include a distribution network self-healing system based on distribution network SCADA, specifically:

[0158] The data collection module collects data and performs preprocessing according to the SCADA system.

[0159] The anomaly detection module extracts features from the preprocessed data and uses the optimized LOF algorithm to detect anomalies.

[0160] The self-healing module isolates faults, reconstructs the grid topology, and realizes self-healing of the distribution network.

[0161] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of a power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method is implemented.

[0162] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0163] Example 2, an embodiment of the present invention, provides a distribution network self-healing method and system based on distribution network SCADA. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0164] The test object selected a distribution network containing multiple lines and load points. The distribution network is equipped with a SCADA system that can perform real-time data collection and monitoring.

[0165] The test equipment includes voltage sensors, current sensors, frequency sensors, power sensors, phase angle sensors and power quality monitors. All sensors are connected to the SCADA system through a communication network to ensure real-time data collection and transmission.

[0166] Test preparation:

[0167] Test equipment: including voltage sensor, current sensor, frequency sensor, power sensor, phase angle sensor and power quality monitor. All sensors are connected to the SCADA system through a communication network to ensure real-time data collection and transmission.

[0168] Test environment: A distribution network with seven load points and five transmission lines was selected. Each load point was equipped with necessary measurement equipment to ensure that detailed electrical parameters could be collected.

[0169] Fault simulation: During the experiment, faults including short circuit and ground fault are artificially introduced to test the effectiveness of the self-healing method.

[0170] The existing technical steps include: collecting the operation data of the distribution network, including voltage, current, frequency, power, phase angle and power quality, through the traditional monitoring system, and transmitting the data to the central control room through the communication network.

[0171] Use preset thresholds for fault detection. When the voltage or current exceeds the preset threshold, it is judged as a fault. Manually operate switches and circuit breakers to isolate the fault area. Adjust the grid topology through manual operation. It is impossible to combine real-time data and optimization algorithms for dynamic reconstruction.

[0172] The steps of our invention include: the SCADA system collects the operation data of the distribution network in real time through sensors, including voltage, current, frequency, power, phase angle and power quality. The data is transmitted to the central control room of the SCADA system through a communication network. The collected data is preprocessed and standardized by a normalization method to ensure that data of different dimensions can be effectively compared.

[0173] The preprocessed data is subjected to feature extraction to extract the characteristic values ​​of various parameters, such as the maximum, minimum, and mean values ​​of voltage, the RMS and peak values ​​of current, the standard deviation of frequency, the maximum, minimum, and average values ​​of power, the mean value of phase angle, and the total harmonic distortion of power quality.

[0174] The optimized LOF algorithm is used to detect abnormal conditions, and the genetic algorithm is combined to optimize the parameters to ensure the accuracy and efficiency of detection. The LOF algorithm evaluates the degree of abnormality of each data point by calculating the local reachable density (LRD), and determines whether there is a fault based on the comprehensive abnormality score.

[0175] After detecting the fault point, the SCADA system sends instructions to the automatic switches and circuit breakers, and the automatic switches disconnect the faulty equipment and lines, cutting off the power supply to the faulty area.

[0176] According to real-time load data and grid topology, linear programming and ant colony algorithm are used to optimize grid reconstruction scheme, adjust grid topology, and transfer load to healthy power supply path. Ant colony algorithm updates pheromone and heuristic information to find the optimal reconstruction path and ensure balanced load distribution.

[0177] The experimental results are shown in Table 1.

[0178]

[0179]

[0180] Through the table data, we can clearly see the significant advantages of our invention in the process of power grid self-healing. After the fault is handled in the prior art, the voltage at the load point fluctuates greatly, which are 215V, 210V and 220V respectively. The fluctuation range is wide, which can easily lead to damage to power equipment and degradation of power supply quality. Our invention uses a SCADA system for data collection and fault isolation, and the voltage at the load point is kept in a relatively stable range, which are 230V, 228V and 232V respectively, which significantly improves the stability of the voltage and ensures the quality of power supply.

[0181] After the fault is handled in the prior art, the current at the load point fluctuates greatly, which are 100A, 95A and 102A respectively, and the load cannot be effectively balanced. Our invention uses an optimized grid reconstruction solution to evenly distribute the current at the load point, which are 105A, 103A and 106A respectively, ensuring the load consistency of the grid and reducing the loss of lines and equipment. The frequency of the load point in the prior art has a certain deviation, which are 49.8Hz, 49.5Hz and 49.9Hz respectively, which may affect the stability of the power system. Through the application of the optimization algorithm, our invention stabilizes the frequency of the load point at around 50Hz, which are 50.2Hz, 50.1Hz and 50.3Hz respectively, improving the frequency accuracy of the power system and ensuring the stable operation of the system.

[0182] After the fault is handled, the existing technology has uneven distribution of active power and reactive power, which are 20kW, 18kW and 22kW respectively, with low active power, and reactive power of 5kVAR, 4.5kVAR and 5.5kVAR respectively, and the power factor is not high. Our invention improves the power factor and overall power quality through optimized grid reconstruction and load adjustment, with active power of 24kW, 23kW and 25kW respectively, and reactive power of 6kVAR, 5.8kVAR and 6.2kVAR respectively.

[0183] The phase angles of the existing technology are 30°, 28° and 32°, and the power quality (total harmonic distortion THD%) is 5.5%, 5.8% and 5.3%, respectively, which affects the power supply quality and equipment life. Our invention uses feature extraction and optimization algorithms, and the phase angles are 31°, 30° and 33°, respectively, and the power quality is significantly improved, and the THD% is 3.2%, 3.4% and 3.1%, respectively, which improves the power supply quality and extends the service life of the equipment.

[0184] Through analysis, it can be seen that the distribution network self-healing method based on distribution network SCADA invented by our party has achieved efficient self-healing of the power grid through data collection and preprocessing, feature extraction and anomaly detection, fault isolation and power grid reconstruction. Compared with the existing technology, our invention significantly improves the stability of voltage and current, the accuracy of frequency, the degree of power optimization, as well as the phase angle and power quality, ensuring the reliability and quality of power supply. Especially in the process of fault isolation and power grid reconstruction, the application of optimization algorithms improves the response speed and decision-making accuracy, and effectively solves the shortcomings of the existing technology in fault handling.

[0185] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A distribution network self-healing method based on distribution network SCADA, characterized in that: include: Collect data and perform preprocessing according to the SCADA system; Perform feature extraction on the preprocessed data and use the optimized LOF algorithm to detect abnormal situations; Conduct fault isolation, reconstruct the power grid topology, and achieve self-healing of the distribution network.

2. The distribution network self-healing method based on distribution network SCADA as claimed in claim 1, characterized in that: The data collection and preprocessing according to the SCADA system include: the SCADA system collects the operation data of the distribution network in real time through sensors and instruments, and transmits the data through a communication network. The collected data includes voltage, current, frequency, power, phase angle and power quality; Data preprocessing includes data normalization, which is expressed as: Among them, x i ′ represents the normalized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

3. The distribution network self-healing method based on distribution network SCADA as claimed in claim 2, characterized in that: The feature extraction of the pre-processed data includes: voltage feature extraction is expressed as: Among them, T V Represents voltage characteristics, V max Indicates the maximum voltage, V min Indicates the minimum voltage, V arg Indicates the average voltage; The current feature extraction is expressed as: Among them, T I Represents the current characteristics, I rms Represents the RMS value of current, I peak Indicates the peak value of the current; The frequency feature extraction formula is expressed as: T f =f std Among them, T f Represents the frequency characteristics, f std represents the standard deviation of the frequency; The power feature extraction is expressed as: Among them, T P Represents the active power characteristic, P max Indicates the maximum active power, P min Indicates that there is a minimum value for the public highway, P arg Indicates the average value of active power; T Q Represents reactive power characteristics, Q max Indicates the maximum reactive power, Q min Indicates the minimum reactive power, Q arg Indicates the average value of reactive power; The phase angle feature extraction is expressed as: T θ =cos(θ avg ) Among them, T θ Represents the phase angle characteristic, θ avg represents the average phase angle; The power quality feature extraction is expressed as: Among them, T PQ Indicates the power quality characteristics, H total Indicates total harmonic distortion, V rms Indicates the RMS value of voltage.

4. The distribution network self-healing method based on distribution network SCADA as claimed in claim 3, characterized in that: The use of the optimized LOF algorithm to detect abnormal situations includes fusing features, which is expressed as: Among them, F i represents the normalized features, T i represents the value of feature i, μ i represents the mean of feature i, σ i represents the standard deviation of feature i; The calculation formula of LOF algorithm is expressed as: Among them, lrd k (p) represents the local reachable density, expressed as: Among them, p represents the point to be detected, N k (p) represents the k nearest neighbor set of point p, reach_dist k (p,o) represents the reachable distance, reach_dist k (p,o)=max{dist(p,o),min_dist k (o)}; dist k (p,o) represents the distance between points p and o, min_dist k (o) represents the kth nearest neighbor distance from point o to its k nearest neighbors; Use genetic algorithms to optimize the calculation process of the LOF algorithm and improve the accuracy and efficiency of the algorithm. The genetic algorithm includes initialization, fitness calculation, selecting some individuals as parents of the next generation according to fitness, performing crossover operations on the selected individuals, generating new individuals, introducing randomness through mutation, and iteration; Initialization includes randomly generating an initial population, each individual represents a local reachable density calculation scheme; the initial population is {C1,C2,...,C N }, each individual Ci contains a parameter set {α i ,β i The fitness calculation includes a formula based on the local anomaly factor to calculate the local reachable density of each individual in the current population and compare it with other individuals. The fitness function is expressed as: Among them, f(C i ) represents the fitness function, Indicates the LOF value calculated based on the Ci parameter; The optimization parameters are obtained by genetic algorithm, α represents the voltage difference adjustment factor determined by genetic algorithm, and β represents the current difference adjustment factor determined by genetic algorithm; the distance calculation formula after optimization of the optimization parameters is expressed as: Among them, dist opt (p,o) represents the physical distance between points p and o, V p Represents the voltage at point p, V o Indicates the voltage at point o, I p represents the current at point p, I o represents the current at point o; The local reachable density formula after optimization of the optimization parameters is expressed as: The optimized local reachable density is used to calculate the LOF value of each point and determine whether it is an abnormal point, which is expressed as: in, represents the optimized reachable distance, 5. The distribution network self-healing method based on distribution network SCADA as claimed in claim 4, characterized in that: The method of detecting abnormal situations by using the optimized LOF algorithm further includes combining the fused features and the optimized LOF algorithm to comprehensively calculate an abnormality score, which is expressed as: Among them, A p represents the comprehensive anomaly score of point p, N represents the number of features, and F i Represents the normalized value of feature i, LOF k (p) represents the anomaly score of point p, λ i represents the weight of feature i; Comprehensive abnormality score A p When >0.5, the point is judged as abnormal; otherwise, it is normal.

6. The distribution network self-healing method based on distribution network SCADA as claimed in claim 5, characterized in that: The fault isolation includes: p >0.5, the fault is isolated, and the SCADA system sends instructions to the automation switch, which disconnects the faulty equipment and lines and cuts off the power supply to the faulty area.

7. The distribution network self-healing method based on distribution network SCADA as claimed in claim 6, characterized in that: The reconstructed power grid topology structure includes a goal of minimizing load imbalance and reconstruction cost. The objective function and constraint conditions of the power grid reconstruction based on linear programming are expressed as: subject to in, represents the load of the i-th line after reconstruction, represents the load of the ith line before reconstruction, C j represents the cost of the jth reconstruction operation, x j A binary variable representing the jth reconstruction operation, 1 indicates that a reconstruction operation is performed, and 0 indicates that no reconstruction operation is performed; Capacity i represents the capacity of the ith line, represents the load of the kth transformer after reconstruction, Capacity k represents the capacity of the kth transformer; The ant colony algorithm is introduced to optimize the linear programming-based power grid reconstruction method, the ant colony is initialized, and the initial pheromone concentration is set; in the power grid topology, each line and transformer is assigned an initial pheromone concentration τ0; In the grid reconstruction problem, the choice of each line and transformer constitutes the solution space, and each ant selects a path based on the pheromone concentration τ and heuristic information η; the selection probability P ij It is expressed as: Among them, τ ij represents the pheromone concentration of line i to j, η ij represents the heuristic information of line i to j, including line capacity and current load; allowed represents the set of optional paths for ants, α' represents the pheromone weight, and β' represents the heuristic information weight; After each iteration, the pheromone concentration is updated according to the path selection of the ants, expressed as: Where ρ represents the pheromone volatility coefficient, represents the amount of pheromone released by the kth ant on the path from i to j; The path selection and pheromone updating process is repeated until a high-quality solution is obtained, and the optimal reconstruction plan is determined, including the switches and circuit breakers that need to be operated, as well as the load transfer path.

8. A distribution network self-healing system based on distribution network SCADA using the method according to any one of claims 1 to 7, characterized in that: The data collection module collects data and performs preprocessing according to the SCADA system; The anomaly detection module extracts features from preprocessed data and uses the optimized LOF algorithm to detect anomalies; The self-healing module isolates faults, reconstructs the grid topology, and realizes self-healing of the distribution network.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images