A method for predicting the state of substation equipment based on AI analysis
Through the AI-based substation device state prediction method, sensor data and convolutional neural network are used to analyze the aging situation, generate repair solutions and detect the repair process in real time, solving the problem of inefficiency in the existing technology and improving the failure of substation device Predict and repair quality, saving maintenance staff working time.
Patent Information
- Application Number
- CN202111320217.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-09
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2041-11-09
AI Technical Summary
The existing technology cannot effectively support the maintenance command decision-making and on-site execution efficiency of substation equipment of power grid companies, and maintenance personnel need to manually write repair logs, which is a waste of time.
The AI-based substation device state prediction method is adopted to build a prediction model through sensor collection data, use convolutional neural network to perform aging analysis, generate repair plans and detect repair processes in real time, automatically record repair logs, reduce misjudgments and improve repair quality.
It improves the accuracy and repair efficiency of substation equipment failure prediction, reduces the work burden of maintenance personnel, saves time to write repair logs, and enhances the supportiveness and repair quality of maintenance decisions.
Smart Images

Figure CN114004528B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of equipment intelligent maintenance, and in particular to a method for predicting the state of substation equipment based on AI analysis. Background Art
[0002] A substation is an electric power facility in the power system that transforms voltage, receives and distributes electric energy, controls the flow of electricity, and adjusts voltage. It connects power grids of various voltage levels through its transformers, and various components are required in the substation. These equipment types are numerous, including transformers, switches, four small devices, reactive devices, and other equipment and auxiliary devices, such as surge arresters, insulators, high-voltage bushings, guide wires, grounding devices, secondary equipment, high-voltage DC equipment, etc. These devices are collectively referred to as substation equipment. Substation equipment plays an important role in the power system, so during use or before use, it is necessary to detect faults in the substation equipment. If the substation equipment is in use, it is necessary to perform predictive analysis on the substation equipment first, usually based on the time parameter failure rate of the substation equipment and the health index parameter failure rate of the substation equipment for analysis and prediction; therefore, it is particularly important to invent a substation equipment status prediction method based on AI analysis;
[0003] After searching, Chinese patent number CN110930052A discloses a method, system, device and readable storage medium for predicting the failure rate of substation equipment. Although the invention can achieve accurate prediction of the failure rate of substation equipment, it cannot effectively support the power grid company's substation maintenance command decision-making and on-site execution efficiency; in addition, the existing substation equipment status prediction method based on AI analysis requires maintenance personnel to write repair logs by themselves, which wastes the maintenance personnel's time; for this reason, we propose a substation equipment status prediction method based on AI analysis. Summary of the invention
[0004] The purpose of the present invention is to solve the defects in the prior art and to propose a method for predicting the state of substation equipment based on AI analysis.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] A method for predicting the state of substation equipment based on AI analysis, the specific steps of the prediction method are as follows:
[0007] (1) Collecting various information about substation equipment to build a prediction model: Each group of sensors detects the substation equipment and transmits each group of data to AI. AI builds a prediction model on its own, receives relevant data, and classifies and processes it. At the same time, it imports each group of classified data into the prediction model;
[0008] (2) Analyze and judge each component of the substation equipment: collect information on each component in the substation equipment and import it into the prediction model to predict the failure of each component and perform data screening on the prediction results;
[0009] (3) Construct a convolutional neural network to analyze the aging of equipment: AI constructs the corresponding convolutional neural network by itself, and trains the weights through the convolutional neural network. The prediction model is updated with the trained weights, and the aging of the substation equipment circuit is analyzed and predicted.
[0010] (4) Collect various analysis data to generate repair solutions: AI receives various data generated by the prediction model, calls corresponding repair solutions from the repair database to build a solution list, and feeds the solution list back to the staff for selection and repair;
[0011] (5) Collect repair information to perform repair analysis on substation equipment: After the repair of the substation equipment is completed, the sensor re-collects various information of the substation equipment, and AI performs abnormal analysis on its repair status. At the same time, it generates a repair log for storage.
[0012] Furthermore, the specific steps of the classification process in step (1) are as follows:
[0013] Step 1: Classify the collected substation equipment-related data according to equipment information, current data, component information, and weather information;
[0014] Step 2: Clean and standardize the classified data, import each group of processed data into the prediction model, and build a corresponding three-dimensional model of the substation equipment.
[0015] Furthermore, the specific steps of fault prediction in step (2) are as follows:
[0016] Step 1: Arrange the component information of each group according to the first letters a to z, and calculate the failure probability of each component using the Bayesian algorithm according to the arrangement order. The specific algorithm formula is as follows:
[0017]
[0018] Among them, P(B i ) represents the basic probability of component failure, P(A|B i ) represents the failure rate, P(A|B j ) represents the false alarm rate;
[0019] Step 2: Based on the calculation results, mark the faulty components at the corresponding positions of the 3D model, generate a fault record table, and record the component name, component location, and fault cause;
[0020] Step 3: Comprehensively consider the reliability of fault probability and the completeness of historical fault information, and determine the fault distribution matrix through information fusion method. At the same time, screen the fault record table and eliminate misdiagnosed data.
[0021] Furthermore, the specific steps of analysis and prediction in step (3) are as follows:
[0022] S1.1: Extract historical data from the cloud database, perform consistency checks based on it, and build a quantitative prediction algorithm model;
[0023] S1.2: import the current information into the algorithm model, extract the abnormal current data, and mark the corresponding circuit position in the three-dimensional model;
[0024] S1.2: The circuit information, circuit location and aging degree are entered into the fault record table.
[0025] Furthermore, the specific steps of selecting and repairing in step (4) are as follows:
[0026] S2.1: The staff checks the list of solutions and selects a repair solution from the list or develops a repair solution on their own;
[0027] S2.2: AI receives the user selection information and sets the repair plan selected by the user as the default plan. If the user makes a repair plan by himself, the repair plan is imported into the plan list and set as the default plan.
[0028] S2.3: AI repairs the faults of the substation equipment according to the corresponding repair plan and fault record table.
[0029] Furthermore, the specific steps of abnormal analysis in step (5) are as follows:
[0030] P1: AI detects the repair process in real time and collects abnormal repair information. It also inspects, operates, and shuts down the repaired substation equipment.
[0031] P2: Collect basic information of substation equipment in various operations and analyze whether there is any abnormal operation information;
[0032] P3: If it exists, locate the location where it occurs, and feed it back to the maintenance personnel in the corresponding area together with the abnormal repair information, and prompt the maintenance personnel to repair it in time. At the same time, generate a repair log to record the repair date, maintenance personnel name and repair content.
[0033] Compared with the prior art, the present invention has the following beneficial effects:
[0034] 1. Compared with the previous manual analysis of substation equipment failure, the substation equipment status prediction method based on AI analysis collects various basic information of substation equipment through AI, and constructs prediction models and three-dimensional models by itself. At the same time, it collects and records the information of faulty components and aging circuits in the substation equipment, eliminates the misjudged data, and calls the corresponding repair scheme from the repair database to build a scheme list. At the same time, the scheme list is fed back to the staff for selection and repair, which can effectively support the power grid company's substation maintenance command decision-making and on-site execution efficiency, and improve the work efficiency of the staff;
[0035] 2. In the method for predicting the state of substation equipment based on AI analysis, AI performs real-time detection of the repair process and collects abnormal repair information. At the same time, it inspects, operates and shuts down the repaired substation equipment. During the operation, it records the basic information of each group of substation equipment and analyzes whether there is abnormal operation information. If so, it locates the location where it occurs and feeds it back to the maintenance personnel in the corresponding area together with the abnormal repair information, prompting the maintenance personnel to repair it in time. At the same time, it generates a repair log to record relevant information and uploads the completed repair log to the storage cloud for storage. It can detect unrepaired vulnerabilities and improve the repair quality of substation equipment. At the same time, there is no need for maintenance personnel to write repair logs by themselves, saving maintenance personnel time. 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 method for predicting the state of substation equipment based on AI analysis proposed by the present invention. DETAILED DESCRIPTION
[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely 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 is necessary to understand that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the 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 orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.
[0040] Example 1
[0041] Reference Figure 1 , a method for predicting the state of a substation equipment based on AI analysis, this embodiment mainly discloses an analysis and prediction method:
[0042] Collect various information of substation equipment to build a prediction model: each group of sensors detects the substation equipment and transmits each group of data to AI. AI builds the prediction model by itself, receives relevant data, and classifies it. At the same time, it imports each group of classified data into the prediction model.
[0043] Specifically, the collected data related to the substation equipment are classified according to equipment information, current data, component information and weather information, and then the classified data are cleaned and standardized. The processed groups of data are imported into the prediction model, and the corresponding three-dimensional model of the substation equipment is constructed.
[0044] It should be further explained that after the three-dimensional model is constructed, the information of each component of the substation equipment will be marked in the three-dimensional model and fed back to the staff for review and adjustment.
[0045] Analyze and judge each component of the substation equipment: collect information on each component in the substation equipment, import it into the prediction model, predict faults for each component, and perform data screening on the prediction results.
[0046] Specifically, AI arranges the information of each group of components according to the first letters a to z, and calculates the fault probability of each component through the Bayesian algorithm according to the arrangement order. After the calculation is completed, the faulty component is marked at the corresponding position of the three-dimensional model based on the calculation results, and a fault record table is generated at the same time. The component name, component location and cause of the fault are recorded. The reliability of the fault probability and the completeness of the historical fault information are comprehensively considered, and the fault distribution matrix is determined through the information fusion method. At the same time, the fault record table is screened to eliminate the misdiagnosed data.
[0047] It needs to be further explained that the specific Bayesian algorithm formula is as follows:
[0048]
[0049] Among them, P(B i ) represents the basic probability of component failure, P(A|B i ) represents the failure rate, P(A|B j ) represents the false alarm rate.
[0050] Construct a convolutional neural network to perform aging analysis on equipment: AI constructs the corresponding convolutional neural network by itself, and trains weights through the convolutional neural network. The prediction model is updated with the trained weights, and the aging of the substation equipment circuit is analyzed and predicted.
[0051] Specifically, AI first extracts historical data from the cloud database and performs consistency verification based on it, and at the same time builds a quantitative prediction algorithm model. After the algorithm model is built, the current information is imported into the algorithm model, the abnormal current data is extracted, and the corresponding circuit position is marked in the three-dimensional model. At the same time, the circuit information, circuit position and aging degree are entered into the fault record table.
[0052] This embodiment uses AI to collect various basic information of substation equipment, and builds prediction models and three-dimensional models by itself. At the same time, it collects and records the information of faulty components and aging circuits in the substation equipment, eliminates misjudged data, and calls the corresponding repair plan from the repair database to build a plan list. At the same time, the plan list is fed back to the staff for selection and repair, which can effectively support the power grid company's substation maintenance command decision-making and on-site execution efficiency, and improve the work efficiency of the staff.
[0053] Example 2
[0054] Reference Figure 1 , a method for predicting the state of substation equipment based on AI analysis, in addition to the same structure as the above embodiment, this embodiment mainly discloses a selective repair method:
[0055] Collect various analysis data to generate repair plans: AI receives various data generated by the prediction model, calls the corresponding repair plans from the repair database to build a plan list, and feeds the plan list back to the staff for selection and repair.
[0056] Specifically, first, the staff checks the plan list and selects a repair plan from the list or makes a repair plan by themselves. After the staff completes the selection, AI receives the user's selection information and sets the repair plan selected by the user as the default plan. If the user makes a repair plan by themselves, the repair plan is imported into the plan list and set as the default plan. AI repairs the substation equipment according to the corresponding repair plan and the fault record table.
[0057] It should be further explained that when the substation equipment fails again, AI will prioritize the corresponding default solution to repair the fault based on the type of fault. At the same time, it will feedback the repair plan to the staff for review, and plot the number and price of components required during the repair process into a price report for storage.
[0058] Collect repair information to perform repair analysis on substation equipment: After the repair of the substation equipment is completed, the sensor re-collects various information of the substation equipment, and AI performs abnormal analysis on its repair status, while generating repair logs for storage.
[0059] Specifically, AI performs real-time detection of the repair process and collects abnormal repair information. At the same time, it inspects, operates, and shuts down the repaired substation equipment. During the operation, it records the basic information of each group of substation equipment and analyzes whether there is abnormal operation information. If so, it locates the location where it occurs and feeds it back to the maintenance personnel in the corresponding area together with the abnormal repair information, prompting the maintenance personnel to repair it in time. At the same time, it generates a repair log to record the repair date, the name of the maintenance personnel, and the repair content, and uploads the completed repair log to the storage cloud for storage.
[0060] This embodiment uses AI to perform real-time detection on the repair process and collect abnormal repair information therein. At the same time, it inspects, operates and shuts down the repaired substation equipment, records basic information of each group of substation equipment during the operation, and analyzes whether there is abnormal operation information. If so, it locates the location where it occurs, and feeds it back to the maintenance personnel in the corresponding area together with the abnormal repair information, and prompts the maintenance personnel to repair it in time. At the same time, a repair log is generated to record relevant information, and the completed repair log is uploaded to the storage cloud for storage. It can detect unrepaired vulnerabilities and improve the repair quality of substation equipment. At the same time, there is no need for maintenance personnel to write repair logs by themselves, saving maintenance personnel time.
[0061] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A method for predicting the state of substation equipment based on AI analysis, characterized in that: The specific steps of the prediction method are as follows: (1) Collecting various information of substation equipment to build a prediction model: Each group of sensors detects the substation equipment and transmits each group of data to AI. AI builds a prediction model on its own, receives relevant data, and classifies and processes it. At the same time, it imports each group of classified data into the prediction model; (2) Analyze and judge each component of the substation equipment: collect information on each component in the substation equipment and import it into the prediction model to predict the failure of each component and perform data screening on the prediction results; (3) Construct a convolutional neural network to analyze the aging of equipment: AI constructs the corresponding convolutional neural network by itself, and trains the weights through the convolutional neural network. The prediction model is updated with the trained weights, and the aging of the substation equipment circuit is analyzed and predicted. (4) Collect various analysis data to generate repair solutions: AI receives various data generated by the prediction model, calls corresponding repair solutions from the repair database to build a solution list, and feeds the solution list back to the staff for selection and repair; (5) Collecting repair information to perform repair analysis on substation equipment: After the substation equipment is repaired, the sensor collects various information about the substation equipment again, and AI performs abnormal analysis on its repair status. At the same time, it generates a repair log for storage. The specific steps of the classification process described in step (1) are as follows: Step 1: Classify the collected substation equipment-related data according to equipment information, current data, component information, and weather information; Step 2: Clean and standardize the classified data, import each group of processed data into the prediction model, and build a corresponding three-dimensional model of the substation equipment; The specific steps of fault prediction described in step (2) are as follows: Step 1: Arrange the component information of each group according to the first letters a to z, and calculate the failure probability of each component using the Bayesian algorithm according to the arrangement order. The specific algorithm formula is as follows: Among them, P(B i ) represents the basic probability of component failure, P(A|B j ) represents the failure rate, P(A|B j ) represents the false alarm rate; Step 2: Based on the calculation results, mark the faulty components at the corresponding positions of the 3D model, generate a fault record table, and record the component name, component location, and fault cause; Step 3: Comprehensively consider the reliability of fault probability and the completeness of historical fault information, and determine the fault distribution matrix through information fusion method. At the same time, screen the fault record table and eliminate the misdiagnosed data. The specific steps of analysis and prediction described in step (3) are as follows: S1.1: Extract historical data from the cloud database, perform consistency checks based on it, and build a quantitative prediction algorithm model; S1.2: Import the current information into the algorithm model, extract the abnormal current data, and mark the corresponding circuit position in the three-dimensional model; S1.2: The circuit information, circuit location and aging degree are entered into the fault record table.
2. The method for predicting the state of substation equipment based on A1 analysis according to claim 1 is characterized in that ,The specific steps of selecting repair in step (4) are as follows: S2.1: The staff checks the list of solutions and selects a repair solution from the list or develops a repair solution on their own; S2.2: A1 receives the user selection information and sets the repair plan selected by the user as the default plan. If the user makes a repair plan by himself, the repair plan is imported into the plan list and set as the default plan. S2.3: AI repairs the substation equipment faults according to the corresponding repair plan and fault record table.
3. According to claim 1, a method for predicting the state of substation equipment based on AI analysis is characterized in that ,The specific steps of abnormal analysis in step (5) are as follows: P1: AI detects the repair process in real time and collects abnormal repair information. It also inspects, operates, and shuts down the repaired substation equipment. P2: Collect basic information of substation equipment in various operations and analyze whether there is any abnormal operation information; P3: If it exists, locate the location where it occurs, and feed it back to the maintenance personnel in the corresponding area together with the abnormal repair information, and prompt the maintenance personnel to repair it in time. At the same time, generate a repair log to record the repair date, maintenance personnel's name and repair content.
