Method, medium and system for shortening fault studying and judging time of power distribution network

Abstract

The invention provides a method, medium and system for shortening power distribution network fault studying and judging time, and belongs to the technical field of power distribution network fault recognition. Comprising the following steps: S1, obtaining power distribution network fault information which comprises a fault alarm signal monitored by a power distribution network operation monitoring system and power distribution network operation state data obtained according to the fault alarm signal; s2, determining a power failure range of the power distribution network according to the fault information; s3, determining a fault type according to the operation state data of the power distribution network; s4, determining a fault power supply line based on the corresponding table of the power failure range and the power supply line; and S5, fault equipment is judged according to the fault information, the power failure range, the fault type and the fault power supply line, and is output to operation and maintenance personnel. By establishing the corresponding table of the power failure range and the power supply line, the time for determining the faulty power supply line is shortened, the fault studying and judging time of the power distribution network is shortened, the fault processing efficiency of the power distribution network is improved, and the power supply reliability of customers is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault identification, and in particular to a method, a medium and a system for shortening distribution network fault analysis time. Background Art

[0002] The distribution network refers to a power grid that receives electric energy from the transmission network or regional power plant, distributes it locally or distributes it step by step to various users according to voltage through distribution facilities. It is composed of overhead lines, cables, poles, distribution transformers, disconnectors, reactive power compensators and some ancillary facilities, and plays an important role in distributing electric energy in the power grid. The distribution network is at the end of the power system, involving a wide range of users and a large number of users. It has extremely high requirements for distribution network dispatching. After a distribution network failure occurs, it is extremely important to quickly handle the failure and restore normal power supply to users. Therefore, how to shorten the fault analysis time of the distribution network and timely discover and locate the fault is an urgent problem to be solved. However, due to the continuous planning of the distribution network, the power supply lines have increased or decreased abnormalities, which makes it impossible to timely determine the faulty power supply lines and faulty equipment, the distribution network fault analysis time becomes longer, and the reliability of customer power supply is reduced. Summary of the invention

[0003] In view of this, the present invention provides a method, a medium and a system for shortening the fault analysis time of a distribution network, so as to solve the technical problems of long fault analysis time and reduced power supply reliability for customers.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] A method for shortening the time for analyzing and judging distribution network faults comprises the following steps:

[0006] S1. Obtain distribution network fault information, wherein the fault information includes a fault alarm signal monitored by a distribution network operation monitoring system and distribution network operation status data obtained according to the fault alarm signal;

[0007] S2. Determine the power outage scope of the distribution network according to the fault information;

[0008] S3, determining the fault type according to the distribution network operation status data;

[0009] S4. Determine the faulty power supply line based on the correspondence table between the power outage range and the power supply line;

[0010] S5. Determine the faulty equipment based on the fault information, power outage scope, fault type and faulty power supply line, and output it to the operation and maintenance personnel.

[0011] Preferably, in step S2, determining the power outage range of the distribution network according to the fault information includes the following steps: comparing the voltage and current data before and after the fault to confirm the parameter changes caused by the fault, and determining the power outage area range through connectivity analysis.

[0012] Preferably, in step S3, determining the fault type according to the distribution network operation status data includes the following steps:

[0013] S31, comparing the power parameter changes before and after the fault;

[0014] S32, obtaining and analyzing current and voltage parameters of the distribution network;

[0015] S33. Determine the fault type according to the change of power parameters, as well as current and voltage parameters.

[0016] Preferably, in step S4, the method of preparing the correspondence table between the power outage range and the power supply line includes:

[0017] Electronic information record of power outages in distribution networks;

[0018] Based on the power outage information ledger, prepare a correspondence table between the distribution network power outage scope and the power supply lines.

[0019] Preferably, the method further comprises: timely updating the correspondence table between the power outage scope and the power supply lines according to the development and changes of the distribution network and the abnormality of the lines.

[0020] Preferably, in step S5, the faulty equipment is judged according to the fault information, power outage scope, fault type and faulty power supply line, and outputted to the operation and maintenance personnel, including: inputting the fault information, power outage scope, fault type and faulty power supply line into a pre-trained fault analysis model, obtaining a list of faulty equipment, and outputting it to the operation and maintenance personnel.

[0021] Preferably, the steps of establishing and training the fault analysis model include:

[0022] Collect historical fault information, power outage scope, fault type, fault power supply line, and historical fault equipment list as training data;

[0023] The LSTM neural network is used to build a time series fault diagnosis model, and the training data is used to train the fault analysis model.

[0024] Preferably, the distribution network operation monitoring system includes multiple data interfaces for connecting and integrating information systems such as distribution automation systems, power regulation and management systems, energy management systems, electricity consumption collection information systems, marketing systems, and production management systems.

[0025] The present invention also provides a computer-readable storage medium, in which program instructions are stored. When the program instructions are executed, the method for shortening the fault analysis time of the distribution network is executed.

[0026] The present invention also provides a system for shortening the time for analyzing and judging distribution network faults, comprising the above-mentioned computer-readable storage medium.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] When the distribution network operation monitoring system detects a fault alarm signal, it indicates that the current distribution network has a fault, and the distribution network operation status data can be obtained at this time. Next, the distribution network power outage range is determined according to the fault information, so that when handling the fault, the equipment and power supply lines can be shut down according to the power outage range to ensure the safety of maintenance personnel. Then, the fault type is determined according to the distribution network operation status data to determine the fault handling method; based on the corresponding table of the power outage range and the power supply line, the power supply line corresponding to the above-determined power outage range is searched, that is, the faulty power supply line; the faulty equipment is judged according to the fault information, power outage range, fault type and faulty power supply line, and output to the operation and maintenance personnel. By establishing a corresponding table of the power outage range and the power supply line, the present invention can timely determine the faulty power supply line, thereby shortening the distribution network fault analysis time, improving the distribution network fault handling efficiency, and improving the customer power supply reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0030] The technical solutions and technical effects of the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings of the present invention.

[0031] Please see Figure 1 A method for shortening the time of fault analysis and judgment of a distribution network comprises the following steps:

[0032] S1. Obtain distribution network fault information, wherein the fault information includes a fault alarm signal monitored by a distribution network operation monitoring system and distribution network operation status data obtained according to the fault alarm signal;

[0033] S2. Determine the power outage scope of the distribution network according to the fault information;

[0034] S3, determining the fault type according to the distribution network operation status data;

[0035] S4. Determine the faulty power supply line based on the correspondence table between the power outage range and the power supply line;

[0036] S5. Determine the faulty equipment based on the fault information, power outage scope, fault type and faulty power supply line, and output it to the operation and maintenance personnel.

[0037] When the distribution network operation monitoring system detects a fault alarm signal, it indicates that the current distribution network has a fault, and the distribution network operation status data can be obtained at this time. Next, the distribution network power outage range is determined according to the fault information, so that when handling the fault, the equipment and power supply lines can be shut down according to the power outage range to ensure the safety of maintenance personnel. Then, the fault type is determined according to the distribution network operation status data to determine the fault handling method; based on the corresponding table of the power outage range and the power supply line, the power supply line corresponding to the above-determined power outage range is searched, that is, the faulty power supply line; the faulty equipment is judged according to the fault information, power outage range, fault type and faulty power supply line, and output to the operation and maintenance personnel. By establishing a corresponding table of the power outage range and the power supply line, the present invention can timely determine the faulty power supply line, thereby shortening the distribution network fault analysis time, improving the distribution network fault handling efficiency, and improving the customer power supply reliability.

[0038] In some implementations, the operation status data of the power grid can be obtained through leakage protection, intelligent switches, fault indicators, power consumption information collection systems, distribution automation and dispatching automation in the system.

[0039] Furthermore, the distribution network operation monitoring system includes multiple data interfaces for connecting and integrating information systems such as distribution automation systems, power control management systems, energy management systems, power consumption information collection systems, marketing systems, and production management systems. The information of multiple information systems is shared with the distribution network operation monitoring system, so that when determining the equipment information, operation data, and power outage information of each device in the distribution network, there is no need to check each system separately, and the data interface corresponding to the information system in the distribution network operation monitoring system is directly called, thereby improving the efficiency of obtaining the distribution network operation status data, thereby improving the efficiency of the subsequent fault analysis process, and shortening the distribution network fault analysis time.

[0040] In the specific implementation process, in step S2, determining the power outage range of the distribution network according to the fault information includes the following steps: comparing the voltage and current data before and after the fault to confirm the parameter changes caused by the fault, and determining the power outage area range through connectivity analysis. Specifically, the distribution network operation status data of the distribution network operation detection system is read, and software such as MATLAB is called for analysis. Taking the distribution transformer data as an example, the voltage and current sampling data of the monitoring point 5 minutes before and after the fault is extracted, and curve comparison and parameter calculation are performed. If the voltage drops significantly, such as dropping to less than 80% of the rated value, the monitoring point is marked. According to the situation of the distribution system and the electrical topology of the equipment, the power supply line of the monitoring point and the power supply line associated with it are determined, and the determination results of multiple monitoring points are combined to obtain the range of the power outage area. Among them, the situation of the distribution system includes the structure of the distribution system, the main equipment, the power supply, etc. The electrical topology of the equipment includes the electrical connection relationship between the various devices, the electrical parameters of the various devices, etc. The power outage range is determined by the situation of the distribution system and the electrical topology of the equipment to avoid misoperation and accidents caused by unclear power outage range.

[0041] Furthermore, maintenance will have a certain impact on surrounding equipment and areas, so the impact range needs to be considered. The size of the impact range depends on the scale of the maintenance equipment and the connectivity of the surrounding equipment. The impact range can be determined by viewing the layout of surrounding related equipment and the parameters of the maintenance equipment. The power outage range can be optimized through the impact range to obtain an accurate power outage range and reduce the impact of power outage maintenance on surrounding users.

[0042] In the specific implementation process, in step S3, determining the fault type according to the distribution network operation status data includes the following steps:

[0043] S31, comparing the power parameter changes before and after the fault;

[0044] Compare the active power before and after the main transformer. If the change exceeds the threshold, it is determined that there is a sudden power change fault. Specifically, the active power parameters of the main station or important area station within 5 minutes before and after the fault are extracted from the monitoring data of the fault analysis system, and curves are drawn and numerical comparisons are performed. Among them, the sampling rate of the power parameter is selected as once every 10 seconds. If there is a sharp increase or decrease in power, that is, ΔP / Δt exceeds the preset threshold, it is preliminarily determined that a short circuit fault has occurred. On the contrary, if the power change is not large, proceed to the next step of parameter analysis. The power fluctuation and threshold range of the short-circuit fault judged above can be used as the corresponding judgment basis based on the changes in historical short-circuit faults and monitoring data.

[0045] S32, obtaining and analyzing current and voltage parameters of the distribution network;

[0046] The fluctuation characteristics of parameters such as voltage and current in the monitoring system can be analyzed by spectrum analysis to extract characteristic frequency components to determine the type of fault. Specifically, for the current waveform data before and after the fault, methods such as wavelet transform can be used to extract and analyze features to determine whether there is an abnormal increase in harmonics, pulses, etc., so as to infer the possible type of fault. For example, if the voltage Kerr effect occurs, it may be a high-resistance grounding fault; if the voltage fluctuates but the current is stable, it is a voltage drop fault; if the current and voltage are distorted at the same time, it may be a line disconnection condition.

[0047] S33. Determine the fault type according to the change of power parameters, as well as current and voltage parameters.

[0048] According to the above power parameter analysis, if it meets the characteristics of a short-circuit fault, that is, the current and voltage change dramatically at the same time, the fault type is determined to be a short-circuit fault. If the fluctuation is not large, and after comprehensive analysis, it meets the characteristics of load changes, it is determined to be a load fault. Furthermore, the duration of the fault can be counted. If the current and voltage return to normal within 1 minute, it can be determined to be a self-recovering fault. The above analysis can use the historical power parameter changes when the fault occurs to set the judgment of drastic changes, the size of the load change, and the range of fluctuations vertically.

[0049] Furthermore, in step S4, the method of preparing the correspondence table between the power outage range and the power supply line includes:

[0050] The electronic information record of power outages in the distribution network; 10kV distribution lines supplied by substations of various voltage levels under the jurisdiction of the power supply station, and each 10kV line is classified according to the equipment switches to summarize important customers such as streets, schools and hospitals, and the electronic information record of power outages in the distribution network is prepared according to the summary table of power supply customers of the distribution network lines. Among them, the electronic information record of power outages in the distribution network includes information such as power supply unit, power outage type, power outage area (equipment), power outage scope, power outage information status, power outage plan time, power outage reason, on-site power supply type, power outage change time, on-site power supply time, release channel, power outage affected equipment (transformer) list, power outage affected user list, etc.

[0051] In some embodiments, the power outage type can be filled in according to the power outage classification, mainly including planned power outage, other temporary power outage, temporary power outage before the power outage, power outage due to power grid fault exceeding the power grid power supply capacity, planned power outage due to power grid fault exceeding the power grid power supply capacity, temporary power outage due to power grid fault exceeding the power grid power supply capacity, etc. The power outage area (equipment) includes the power supply facilities (equipment) involved in the power outage, that is, the name of the power supply facility, the number of the power supply facility, the transformer attributes (public transformer / special transformer), etc. The power outage scope includes the geographical location of the power outage, the high-risk and important customers involved, the special transformer customers, hospitals, schools, towns (streets), villages (communities), residential areas, etc. Information. Power outage information status: divided into two categories: valid and invalid. The planned power outage time includes the start time and expected end time of the planned power outage, temporary power outage, and power outage exceeding the power grid power supply capacity, and the fault power outage includes the fault start time and the expected fault repair time. The power outage change time includes the changed power outage plan start time and the planned power supply time, including new additions, planned changes, temporary changes, and cancellations. The list of equipment (transformers) affected by the power outage includes equipment name, equipment type, equipment identification, etc. The scope of transformers includes public and special transformers that supply power to normal electricity users. Equipment types: including transformers, high-voltage motors, etc. List of users affected by power outages: including user names, user codes, equipment identification, etc. The scope of users includes normal electricity users, excluding users who are under assessment or have their accounts cancelled.

[0052] According to the power outage information ledger, a corresponding table between the power outage range and the power supply line of the distribution network is compiled. Specifically, by extracting the power outage range in the power outage information ledger and the corresponding power supply line, a corresponding table between the power outage range and the power supply line of the distribution network is obtained, so that the power supply line can be directly determined from the power outage range according to the corresponding table, improving the efficiency of determining the faulty power supply line, and thus shortening the time for fault analysis of the distribution network.

[0053] Furthermore, according to the development and changes of the distribution network and the changes in the lines, the corresponding table of power outage scope and power supply lines is updated in a timely manner. Through the OMS system, the statistics of distribution network line changes are queried on a monthly basis to understand the operating status of the power supply lines. The new or reduced power supply areas of the distribution network are added or subtracted in a timely manner in the corresponding table, and the existing corresponding table is updated.

[0054] Further, in step S5, the faulty equipment is judged according to the fault information, power outage scope, fault type and faulty power supply line, and outputted to the operation and maintenance personnel, including: inputting the fault information, power outage scope, fault type and faulty power supply line into a pre-trained fault analysis model, obtaining a list of faulty equipment, and outputting it to the operation and maintenance personnel.

[0055] Furthermore, the steps of establishing and training the fault analysis model include:

[0056] Collect historical fault information, power outage scope, fault type, fault power supply line, and historical fault equipment list as training data;

[0057] The LSTM neural network is used to build a time series fault diagnosis model, and the training data is used to train the fault analysis model.

[0058] Specifically, historical fault information, power outage scope, fault type, and faulty power supply line are collected, the faulty equipment category is marked, and the equipment operating parameters are used as input features. A convolutional neural network (CNN) model is used to set up structures such as convolutional layers and pooling layers to train the faulty equipment diagnosis model. To improve the accuracy of the model, the amount of training data is expanded and the network hyperparameters are adjusted. The relevant equipment parameters of this fault are encoded and formatted. The processed information is formed into image or vector data and input into the trained CNN model. The CNN model outputs the equipment category judgment result corresponding to this fault. The comprehensive model judges the confidence and outputs the most likely faulty equipment.

[0059] Use LSTM neural network to build fault equipment diagnosis model. The collected input feature vectors include: X = [U, I, P, T run , Θ, Φ], where Θ and Φ are the voltage phase angle and current phase angle before and after segmentation, respectively.

[0060] The specific structure of the LSTM neural network is as follows: the network input is the multi-dimensional time series data including voltage, current, phase, etc. obtained by this solution. The LSTM layer learns the time-dependent characteristics of the input data by controlling the input gate, forget gate and output gate inside the control unit. Finally, the output layer performs multi-classification prediction of the faulty device category for the current input and gives the probability of each possible device. The model judgment performance is improved through transfer learning. The LSTM network uses the time series information of the power parameters to accurately predict the dynamic change process of the faulty equipment.

[0061] After training, the LSTM model can be deployed online. When a new fault sample arrives, its feature vector is extracted and the model generates the classification output probability of the faulty device at the current moment. The results at multiple moments are counted and the most likely faulty device is output.

[0062] Furthermore, the present invention also provides a computer-readable storage medium, in which program instructions are stored. When the program instructions are executed, they are used to execute the above-mentioned method of shortening the fault analysis time of the distribution network.

[0063] Furthermore, the present invention also provides a system for shortening the fault analysis time of a distribution network, comprising the above-mentioned computer-readable storage medium.

[0064] What is disclosed above is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiments and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A method for shortening the time for analyzing and judging distribution network faults, characterized in that: The following steps are involved: S1. Obtain distribution network fault information, wherein the fault information includes a fault alarm signal monitored by a distribution network operation monitoring system and distribution network operation status data obtained according to the fault alarm signal; S2. Determine the power outage scope of the distribution network according to the fault information; S3, determining the fault type according to the distribution network operation status data; S4. Determine the faulty power supply line based on the correspondence table between the power outage range and the power supply line; S5. Determine the faulty equipment based on the fault information, power outage scope, fault type and faulty power supply line, and output it to the operation and maintenance personnel.

2. The method for shortening the time for analyzing and judging distribution network faults according to claim 1 is characterized in that: In step S2, determining the power outage range of the distribution network according to the fault information includes the following steps: comparing the voltage and current data before and after the fault to confirm the parameter changes caused by the fault, and determining the power outage area range through connectivity analysis.

3. The method for shortening the time for analyzing and judging distribution network faults according to claim 1 is characterized in that: In step S3, determining the fault type according to the distribution network operation status data includes the following steps: S31, comparing the power parameter changes before and after the fault; S32, obtaining and analyzing current and voltage parameters of the distribution network; S33. Determine the fault type according to the change of power parameters, as well as current and voltage parameters.

4. The method for shortening the time for analyzing and judging distribution network faults according to claim 1, characterized in that: In step S4, the method of preparing the correspondence table between the power outage range and the power supply line includes: Prepare electronic information records of power outages in distribution networks; Based on the power outage information ledger, prepare a correspondence table between the distribution network power outage scope and the power supply lines.

5. The method for shortening the time for analyzing and judging distribution network faults according to claim 4 is characterized in that: The method also includes: timely updating the correspondence table between the power outage range and the power supply line according to the development and changes of the distribution network and the abnormal changes of the lines.

6. The method for shortening the time for analyzing and judging distribution network faults according to claim 1 is characterized in that: In step S5, the faulty equipment is judged according to the fault information, power outage scope, fault type and faulty power supply line, and outputted to the operation and maintenance personnel, including: inputting the fault information, power outage scope, fault type and faulty power supply line into a pre-trained fault analysis model, obtaining a list of faulty equipment, and outputting it to the operation and maintenance personnel.

7. The method for shortening the time for analyzing and judging distribution network faults according to claim 6 is characterized in that: The steps of establishing and training the fault analysis model include: Collect historical fault information, power outage scope, fault type, fault power supply line, and historical fault equipment list as training data; The LSTM neural network is used to build a time series fault diagnosis model, and the training data is used to train the fault analysis model.

8. The method for shortening the time for analyzing and judging distribution network faults according to claim 1, characterized in that: The distribution network operation monitoring system includes multiple data interfaces for connecting and integrating information systems such as distribution automation system, power regulation and management system, energy management system, electricity consumption collection information system, marketing system, and production management system.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the method for shortening the fault analysis time of the distribution network as described in any one of claims 1 to 8.

10. A system for shortening the time of fault analysis and judgment in a distribution network, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.

Images