A power grid operation and maintenance repair method based on log model

Abstract

The present invention discloses a power grid operation and maintenance repair method based on a log model, which belongs to the technical field of power grid operation and maintenance. The specific steps of the repair method are as follows: (1) receiving log data and classifying and processing the log data; (2) establishing a fault knowledge base and extracting information from the fault log; (3) matching the log data by itself and recommending corresponding operation and maintenance solutions; (4) updating the solution list according to the staff; (5) receiving user feedback information and updating the relevant operation and maintenance solutions. The present invention can extract keywords from the log and perform keyword matching, improve the log matching rate, improve the computer data processing efficiency, and update the fault knowledge base in real time, improve the staff's user experience, improve the staff's fault processing rate, save the staff's maintenance time, and at the same time be able to optimize and update according to user needs and improve the solution fault tolerance rate.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation and maintenance, and in particular to a power grid operation and maintenance repair method based on a log model. Background Art

[0002] The power grid, consisting of substations and transmission and distribution lines of various voltages, is a power system. It comprises three units: transformation, transmission, and distribution. The power grid's mission is to transmit and distribute electrical energy and change voltage. Power operation and maintenance (O&M) involves performing necessary maintenance and repairs on power equipment to promptly detect faults, maintain them in good condition, and ensure safe and stable operation of the grid. Therefore, developing a log-based O&M and repair method for power grids is crucial.

[0003] After searching, Chinese patent number CN113472068A discloses a remote operation and maintenance method, system, and storage medium for an island microgrid. Although this invention can realize remote operation and maintenance of the island microgrid and improve the stability of the island microgrid operation, it cannot provide operation and maintenance solutions for staff members, which reduces the staff's grid repair efficiency and wastes staff time. In addition, the existing grid operation and maintenance repair method based on the log model cannot efficiently match log data, which reduces the fault processing rate. To this end, we propose a grid operation and maintenance repair method based on the log model. Summary of the Invention

[0004] The purpose of the present invention is to solve the defects in the prior art and to propose a power grid operation and maintenance repair method based on a log model.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A power grid operation and maintenance repair method based on a log model, the specific steps of the repair method are as follows:

[0007] (1) Receiving log data and classifying and processing it: The staff connects the computer to the power grid system, receives the log data sent by the power grid system in real time, and classifies and processes it;

[0008] (2) Establish a fault knowledge base and extract information from fault logs: Establish a fault knowledge base and analyze and extract each type of fault information in the fault logs. At the same time, extract keywords from each fault-related log and build a matching rule base.

[0009] (3) Automatically match log data and recommend corresponding operation and maintenance solutions: Receive the log data subsequently sent by the power grid system and distribute the data of each group of log data according to the matching database. At the same time, the fault knowledge base automatically recommends the existing relevant fault response operation and maintenance solutions;

[0010] (4) Data update of the solution list based on staff: After the relevant fault response solutions are recommended, the user selects data for each group of solutions, and the fault knowledge base updates the recommended solution list based on the staff selection information;

[0011] (5) Receive user feedback and update relevant operation and maintenance plans: Each user provides feedback on the relevant operation and maintenance plans based on their own usage experience. The fault knowledge base receives the feedback data and updates and repairs the relevant operation and maintenance plans based on the feedback data.

[0012] Furthermore, the specific steps of the classification process in step (1) are as follows:

[0013] Step 1: The computer converts the continuously incoming unstructured IT log data into structured IT log data;

[0014] Step 2: Extract information from each set of structured IT log data and classify the extracted information according to timestamp, operation object, operation time, operation location, operation type, and authorization level.

[0015] Furthermore, the specific steps of analysis and extraction in step (2) are as follows:

[0016] Step 1: Collect past fault logs and perform data clustering. The specific steps for data clustering are as follows:

[0017] Ⅰ: Convert the IP addresses, numbers, and percentages corresponding to each group of fault logs into unified labels;

[0018] II: After the conversion is completed, a bag-of-words model is constructed based on each set of fault log information. The bag-of-words model is represented as an M×V matrix, where M is the corpus length and V is the dictionary length.

[0019] III: The bag-of-words model is constructed and the LDA topic model is used for text clustering;

[0020] Step 2: The fault database extracts the fault cause, fault level and fault handling method information from each group of fault logs and records them.

[0021] Furthermore, the specific steps of data allocation in step (3) are as follows:

[0022] S1: Extract keywords from log data and compare them with each group of data in the matching rule library;

[0023] S2: If there is a matching keyword, the fault log corresponding to the keyword is extracted and the existing relevant fault response operation and maintenance solution is recommended;

[0024] S3: If there is no matching keyword, the log data is entered into the fault knowledge base, and the fault knowledge base is updated at the same time.

[0025] Furthermore, the specific steps of updating the data in step (4) are as follows:

[0026] P1: The user views the operation and maintenance plan list, selects an operation and maintenance plan from the list, and records the operation and maintenance plan number N.

[0027] P1: If N=1, the staff member selects the operation and maintenance plan ranked first in the list, and defaults to retaining the plan and executing it;

[0028] P2: If N≠1, it is determined that the staff has selected the Nth-ranked plan, and the Nth-ranked plan is modified to the first-ranked operation and maintenance plan, and the operation and maintenance plan is implemented;

[0029] P4: If the staff does not select an operation and maintenance plan from the plan list, the computer will provide the operation and maintenance fault information to the staff. The staff will perform new troubleshooting operations. The computer will automatically record the new response plan into the response measures corresponding to the operation and maintenance event and rank it first by default.

[0030] Furthermore, the specific steps of updating and repairing in step (5) are as follows:

[0031] M1: Builds a neural network and receives user feedback, likes, votes, and supplementary explanations on the operation and maintenance solutions recommended by the operation and maintenance fault database;

[0032] M2: The neural network generates updated data based on users' likes, votes, and supplementary explanations, and imports it into the fault knowledge base. The fault knowledge base updates the relevant operation and maintenance plans based on the updated data and feeds back the updated information to the staff.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. Compared with the previous method of sequentially comparing and matching log files, the power grid operation and maintenance repair method based on the log model collects past fault logs by computer and performs data clustering processing on them. The fault database extracts the fault cause, fault level and fault handling method information from each group of fault logs and records them. After the recording is completed, the computer extracts keywords from the subsequently transmitted log data and compares them with each group of data in the matching rule library. If there are keywords that match the comparison, the fault log corresponding to the keyword is extracted, and the existing relevant fault response operation and maintenance plan is recommended. If there are no keywords that match the comparison, the log data is entered into the fault knowledge base, and the fault knowledge base is updated at the same time. It can extract keywords in the log and perform keyword matching, thereby improving the log matching rate, improving the computer data processing efficiency, and updating the fault knowledge base in real time, improving the staff's user experience.

[0035] 2. In the power grid operation and maintenance repair method based on the log model, the fault knowledge base provides relevant operation and maintenance plans to the staff based on the fault log information. The user views the operation and maintenance plan list and selects the operation and maintenance plan in the plan list. The list is updated based on the user's selection of plans in the plan list. If the staff does not select an operation and maintenance plan in the plan list, the computer will provide the operation and maintenance fault information to the staff. The staff will perform new troubleshooting operations. The computer will automatically record the new response plan into the response measures corresponding to the operation and maintenance event and rank it first by default. At the same time, it accepts user feedback on the operation and maintenance plans recommended by the operation and maintenance fault library, votes and additional explanations, and updates and optimizes the fault knowledge base, which can improve the staff's fault processing rate and save staff maintenance time. At the same time, it can be optimized and updated according to user needs to improve the plan's fault tolerance. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0037] Figure 1 This is a flowchart of a log model-based power grid operation and maintenance repair method proposed by the present invention. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0039] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0040] Example 1

[0041] Reference Figure 1 , a power grid operation and maintenance repair method based on a log model, this embodiment mainly provides a data distribution method:

[0042] Receive log data and classify and process it: The staff connects the computer to the power grid system, receives the log data sent by the power grid system in real time, and classifies and processes it.

[0043] Specifically, the computer converts the continuously incoming unstructured IT log data into structured IT log data. After the conversion is completed, the computer extracts information from each group of structured IT log data and classifies the extracted information according to timestamp, operation object, operation time, operation location, operation type and authorization level.

[0044] Establish a fault knowledge base and extract information from fault logs: By establishing a fault knowledge base, analyzing and extracting each type of fault information in the fault logs, and extracting keywords from each fault-related log, a matching rule base is constructed.

[0045] Specifically, past fault logs are collected by computer and data clustering is performed on them. The fault database extracts the fault cause, fault level and fault handling method information from each group of fault logs and records them.

[0046] It should be further explained that the clustering process is specifically that the computer converts the IP addresses, numbers, and percentages corresponding to each group of fault logs into a unified label representation. After successful conversion to label form, the computer constructs a bag-of-words model based on the information of each group of fault logs, where the bag-of-words model is represented as an M×V matrix, where M is the corpus length and V is the dictionary length. After the bag-of-words model is constructed, the LDA topic model is used for text clustering.

[0047] Automatically match log data and recommend corresponding operation and maintenance solutions: By receiving the log data subsequently sent by the power grid system and allocating data for each group of log data based on the matching database, the fault knowledge base automatically recommends existing relevant fault response operation and maintenance solutions.

[0048] Specifically, the computer extracts keywords from the log data and compares them with the data in the matching rule library. If there are keywords that match, the fault log corresponding to the keyword is extracted and the existing relevant fault response and maintenance solutions are recommended. If there are no keywords that match, the log data is entered into the fault knowledge library and the fault knowledge library is updated at the same time.

[0049] In this embodiment, by updating the data of the fault knowledge base, keywords in the log can be extracted and keyword matching can be performed, thereby improving the log matching rate, improving the computer data processing efficiency, and updating the fault knowledge base in real time to improve the staff's user experience.

[0050] Example 2

[0051] Reference Figure 1 , a power grid operation and maintenance repair method based on a log model, in addition to the same structure as the above embodiment, this embodiment discloses a data update method:

[0052] Data update of the solution list based on staff: This is completed through the recommendation of relevant fault response solutions. The user selects data for each group of solutions, and the fault knowledge base updates the recommended solution list based on the staff's selection information.

[0053] Specifically, the user checks the operation and maintenance plan list and selects the operation and maintenance plan in the plan list, and records the operation and maintenance plan number N. If N=1, it is judged that the staff has selected the operation and maintenance plan ranked first in the list, and the plan is retained by default and executed. If N≠1, it is judged that the staff has selected the plan ranked N, and the Nth plan is modified to the operation and maintenance plan ranked first, and the operation and maintenance plan is executed, and the operation and maintenance plan ranked first is deleted. If the staff does not select an operation and maintenance plan in the plan list, the computer will provide the operation and maintenance fault information to the staff, and the staff will perform new troubleshooting operations. The computer automatically records the new response plan into the response measures corresponding to the operation and maintenance event, and defaults to the first place.

[0054] Receive user feedback and update relevant operation and maintenance plans: Each user provides feedback on the relevant operation and maintenance plans based on their own usage experience. The fault knowledge base receives the feedback data and updates and repairs the relevant operation and maintenance plans based on the feedback data.

[0055] Specifically, by building a neural network and receiving user feedback on the operation and maintenance solutions recommended by the operation and maintenance fault database, the neural network generates updated data based on the user's votes and supplementary explanations, and imports them into the fault knowledge database. The fault knowledge database updates the relevant operation and maintenance solutions based on the updated data and feeds back the updated information to the staff;

[0056] In this embodiment, by updating and optimizing the fault knowledge base, the fault processing rate of the staff can be improved, and the maintenance time of the staff can be saved. At the same time, it can be optimized and updated according to user needs to improve the fault tolerance of the solution.

[0057] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A power grid operation and maintenance repair method based on a log model, characterized in that: The specific steps of this repair method are as follows: (1) Receive log data and classify it: The staff connects the computer to the power grid system and receives the log data sent by the power grid system in real time, and classifies it; (2) Establish a fault knowledge base and extract information from fault logs: Establish a fault knowledge base and analyze and extract each type of fault information in the fault logs. At the same time, extract keywords from each fault-related log and build a matching rule base. (3) Automatically match log data and recommend corresponding operation and maintenance solutions: Receive the log data subsequently sent by the power grid system and distribute the data of each group of log data according to the matching database. At the same time, the fault knowledge base automatically recommends the existing relevant fault response operation and maintenance solutions; (4) Data update of the solution list based on staff: After the relevant fault response solutions are recommended, the user selects data for each group of solutions, and the fault knowledge base updates the recommended solution list based on the staff selection information; (5) Receive user feedback and update relevant operation and maintenance plans: Each user provides feedback on the relevant operation and maintenance plans based on their own experience. The fault knowledge base receives the feedback data and updates and repairs the relevant operation and maintenance plans based on the feedback data. The specific steps of the classification process described in step (1) are as follows: Step 1: The computer converts the continuously incoming unstructured IT log data into structured IT log data; Step 2: Extract information from each set of structured IT log data and classify the extracted information according to timestamp, operation object, operation time, operation location, operation type, and authorization level; The specific steps of analysis and extraction described in step (2) are as follows: Step 1: Collect past fault logs and perform data clustering. The specific steps for data clustering are as follows: Ⅰ: Convert the IP addresses, numbers, and percentages corresponding to each group of fault logs into a unified label representation; II: After the conversion is completed, a bag-of-words model is constructed based on each set of fault log information. The bag-of-words model is represented as an M×V matrix, where M is the corpus length and V is the dictionary length. III: The bag-of-words model is constructed and the LDA topic model is used for text clustering; Step 2: The fault database extracts the fault cause, fault level, and fault handling method from each set of fault logs and records them; The specific steps of data allocation in step (3) are as follows: S1: Extract keywords from log data and compare them with each group of data in the matching rule library; S2: If there is a matching keyword, the fault log corresponding to the keyword is extracted and the existing relevant fault response operation and maintenance solution is recommended; S3: If there is no matching keyword, the log data is entered into the fault knowledge base, and the fault knowledge base is updated at the same time.

2. A power grid operation and maintenance repair method based on a log model according to claim 1, characterized in that: The specific steps for updating the data in step (4) are as follows: P1: The user views the operation and maintenance plan list, selects an operation and maintenance plan from the list, and records the operation and maintenance plan number N. P1: If N=1, the staff member selects the operation and maintenance plan ranked first in the list, and defaults to retaining the plan and executing it; P2: If N≠1, it is determined that the staff has selected the Nth-ranked plan, and the Nth-ranked plan is modified to the first-ranked operation and maintenance plan, and the operation and maintenance plan is implemented; P4: If the staff does not select an operation and maintenance plan from the plan list, the computer will provide the operation and maintenance fault information to the staff. The staff will perform new troubleshooting operations. The computer will automatically record the new response plan into the response measures corresponding to the operation and maintenance event and rank it first by default.

3. The power grid operation and maintenance repair method based on a log model according to claim 1, characterized in that: The specific steps for updating and repairing in step (5) are as follows: M1: Builds a neural network and receives user feedback, likes, votes, and supplementary explanations on the operation and maintenance solutions recommended by the operation and maintenance fault database; M2: The neural network generates updated data based on users' likes, votes, and supplementary explanations, and imports it into the fault knowledge base. The fault knowledge base updates the relevant operation and maintenance plans based on the updated data and feeds back the updated information to the staff.

Images