A medium and high voltage misalignment monitoring system based on big data analysis

Through the medium and high voltage misalignment monitoring system based on big data analysis, the difficulty of misalignment monitoring of medium and high voltage and total meter meter meter measurement points is solved, and the misalignment sensitivity coefficient adjustment is realized for line units, which improves the timeliness and accuracy of monitoring. , reduces costs.

CN117992889BActive Publication Date: 2024-11-08STATE GRID JIANGSU ELECTRIC POWER CO LTD MARKETING SERVICE CENT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410055107.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-15
Publication Date
2024-11-08
Estimated Expiration
2044-01-15

AI Technical Summary

Technical Problem

The prior art has difficulties in monitoring misalignment of medium and high voltage and total meter meter meter measurement points, resulting in unfair measurement, abnormal electricity bill settlement, and the inability to adjust the monitoring cycle according to the misalignment sensitivity coefficient of the line unit, reducing work efficiency and resource utilization.

Method used

Using a medium and high voltage misalignment monitoring system based on big data analysis, including clock synchronization module, data acquisition module, industrial server, misalignment analysis module, operation and maintenance management module, model building module, cloud platform and collaborative evaluation module, through data acquisition, Model construction and collaborative evaluation, judge the inaccurate phenomenon and adjust the monitoring period according to the inaccurate sensitivity coefficient.

Benefits of technology

The full range of misalignment monitoring of medium and high voltage and total meter measurement points has been achieved, which improves the timeliness and accuracy of monitoring, reduces labor costs, and improves resource utilization efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117992889B_ABST
    Figure CN117992889B_ABST
Patent Text Reader

Abstract

The invention discloses a medium- and high-voltage misalignment monitoring system based on big data analysis, relates to the field of electric power technology, and solves the technical problems of the current lack of misalignment monitoring and analysis in medium- and high-voltage and total metering points, and the inability to adjust the misalignment monitoring period according to the misalignment sensitivity coefficient of the line unit; the system comprises a data acquisition module, an industrial server, a misalignment analysis module and a collaborative evaluation module; the data acquisition module comprises three switches, which are used to collect power data of each line unit in a monitoring area according to a preset period; the industrial server is used to summarize and process the data information collected by the switch, and input the power data of each line unit into a misalignment phenomenon diagnosis model to obtain an output result to determine whether there is a misalignment phenomenon; the collaborative evaluation module is used to evaluate the misalignment sensitivity coefficient SM of each line unit to adaptively adjust the corresponding misalignment monitoring period, thereby improving the misalignment monitoring efficiency and maximizing resource utilization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a medium- and high-voltage misalignment monitoring system based on big data analysis. Background Art

[0002] At present, there are two types of metering equipment in the power system: high-supply high-meter and high-supply low-meter. These two types of equipment are mainly composed of high-voltage metering boxes, large-user load terminals, multi-function meters and cabinets. These devices are mainly used to measure the power consumption of special transformer users and remotely transmit data to the power supply bureau backend to realize remote power reading, terminal billing and working status functions. However, according to the actual use on site, metering inaccuracy often occurs, which can lead to unfair metering and abnormal electricity bill settlement.

[0003] At present, misalignment analysis is only used at low-voltage user-side metering points, and does not cover medium- and high-voltage and total metering points. It is very difficult to determine the abnormal type, location and cause of the abnormality at medium- and high-voltage and substation total metering points, and a large amount of manpower costs are required for investigation, which reduces work efficiency and company benefits. At the same time, it is impossible to adjust the monitoring period according to the misalignment sensitivity coefficient of the line unit, thereby improving the timeliness and accuracy of medium- and high-voltage misalignment monitoring and maximizing resource utilization. Based on the above shortcomings, the present invention proposes a medium- and high-voltage misalignment monitoring system based on big data analysis. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a medium and high voltage misalignment monitoring system based on big data analysis, which is used to perform misalignment monitoring and analysis on medium and high voltage and main metering points, and at the same time can adjust the misalignment monitoring period according to the misalignment sensitivity coefficient of the line unit, thereby improving the timeliness and accuracy of medium and high voltage misalignment monitoring and maximizing resource utilization.

[0005] To achieve the above-mentioned object, a first aspect of the present invention provides a medium and high voltage misalignment monitoring system based on big data analysis, comprising a clock synchronization module, a data acquisition module, an industrial server, a misalignment analysis module, an operation and maintenance management module, a model building module, a cloud platform and a collaborative evaluation module;

[0006] The data acquisition module includes three switches for collecting power data of each line unit in the monitoring area according to a preset period; the power data includes power consumption at the power supply metering point, power consumption at the load metering point, and fixed power loss;

[0007] The industrial server is used to aggregate and process the data information collected by the switch, and transmit the power data of each line unit to the misalignment analysis module; the power data carries a unique line identifier;

[0008] The model building module is used to periodically acquire historical power data of each line unit to build a sample data set to train a misalignment phenomenon diagnosis model;

[0009] The misalignment analysis module is used to input the power data of each line unit into the misalignment phenomenon diagnosis model, obtain the output result, and determine whether there is a misalignment phenomenon;

[0010] When the output result is 1, it is determined that the corresponding line unit has misalignment, and a misalignment signal is generated to the industrial server; when the output result is 0, it is determined that the corresponding line unit is measuring normally;

[0011] When the line unit is misaligned, the industrial server sends a misalignment signal to the operation and maintenance management module; after receiving the misalignment signal, the operation and maintenance management module arranges power personnel to check the load metering points in the line unit one by one to obtain the misalignment data of the load metering points;

[0012] The collaborative evaluation module is used to evaluate the misalignment sensitivity coefficient SM in combination with the metering points of the line unit and the user side distribution, so as to adaptively adjust the corresponding misalignment monitoring period according to the misalignment sensitivity coefficient SM.

[0013] Furthermore, the specific evaluation process of the collaborative evaluation module is as follows:

[0014] For a certain line unit, the number of load metering points of the line unit is counted as L1; the number of user sides accessing the line unit is counted as H1;

[0015] In a preset time period, the average daily power consumption of the line unit is counted as D1; ​​the misalignment event records of the line unit are collected, and the total number of misalignment times of the line unit is counted as C1;

[0016] The misalignment sensitivity coefficient SM of the line unit is calculated using the formula SM=(H1×a1+D1×a2+C1×a3) / (L1×a4), wherein a1, a2, a3, and a4 are all preset coefficient factors.

[0017] Furthermore, the corresponding misalignment monitoring period is adaptively adjusted according to the misalignment sensitivity coefficient SM, specifically: a comparison table of misalignment sensitivity coefficient range and misalignment monitoring period is pre-stored in the cloud platform;

[0018] According to the comparison table, determining a misalignment sensitivity coefficient range that matches the misalignment sensitivity coefficient SM, and then determining a corresponding misalignment monitoring period according to the misalignment sensitivity coefficient range;

[0019] The collaborative evaluation module is used to feed back the adjusted misalignment monitoring cycle to the industrial server; the industrial server is used to generate a data collection instruction to the switch according to the misalignment monitoring cycle.

[0020] Furthermore, the specific collection steps of the data collection module are as follows:

[0021] Based on the medium and high voltage line power supply mode and line unit, the metering points on the medium and high voltage feeder are assigned to the line unit; the line unit includes a power supply metering point and several load metering points;

[0022] The three switches are all connected to the power consumption information collection system, and one of the switches is used to collect the power consumption of the power supply metering point within the week of the day detected by the power consumption information collection system;

[0023] One switch is used to collect the power consumption of each load metering point within the week of the day detected by the power consumption information collection system; one switch is used to collect the fixed power supply loss within the week of the day detected by the power consumption information collection system; the fixed power supply loss includes line loss and metering equipment loss.

[0024] Furthermore, the relationship between the power supply metering point and the load metering point forms a simple tree topology, and the line power supply metering point provides energy input for the next level load metering point.

[0025] Furthermore, the metering points on the medium and high voltage feeder lines include metering points on the outlet side of the substation, metering points corresponding to the tie switches between adjacent lines, and metering points corresponding to all transformers on the corresponding lines.

[0026] Furthermore, the clock synchronization module is connected to the data acquisition module, and a precise timing tag is assigned to the data collected by the data acquisition module to achieve synchronous transmission of the data.

[0027] Furthermore, the specific training steps of the model building module are as follows:

[0028] S1: sorting and extracting sample data of a given time period to form input and output data sets; the input data refers to the normal power data and inaccurate power data of the line unit within a certain time period;

[0029] The output data is whether there is a misalignment phenomenon, which is represented by 0 and 1; 0 means that the line unit measurement is normal, and 1 means that the line unit has a misalignment phenomenon;

[0030] S2: Divide the data set into a training set, a validation set, and a test set according to a preset ratio; the preset ratio includes 2:1:1, 3:1:1, and 4:3:1;

[0031] S3: Constructing a fusion model: The fusion model is a model constructed by combining at least two of the support vector machine, deep convolutional neural network and RBF neural network;

[0032] S4: After normalizing the data of the training set, test set and verification set, the fusion model is trained, tested and verified, and the trained fusion model is marked as the misalignment diagnosis model.

[0033] Furthermore, fusion methods include linear weighted fusion method, cross fusion method, waterfall fusion method, feature fusion method and predictive fusion method.

[0034] Furthermore, the operation and maintenance management module is used to arrange power personnel to conduct key inspections based on the daily working status of the load metering points to obtain the inaccurate data of the load metering points.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The model building module in the present invention is used to periodically acquire the historical power data of each line unit to build a sample data set to train a misalignment phenomenon diagnosis model; the data acquisition module includes three switches, which are used to collect power data of each line unit in the monitoring area according to a preset period; the industrial server is used to summarize and process the data information collected by the switch, and input the power data of each line unit into the misalignment phenomenon diagnosis model to obtain the output result to determine whether there is a misalignment phenomenon; the misalignment data can be analyzed in the whole range to improve the comprehensiveness of misalignment monitoring; at the same time, fewer computing resources are required, making the monitoring process more efficient and reducing costs;

[0037] 2. The collaborative evaluation module in the present invention is used to evaluate the misalignment sensitivity coefficient SM of each line unit, so as to adjust the misalignment monitoring period according to the misalignment sensitivity coefficient SM; the industrial server is used to generate data acquisition instructions to the switch according to the misalignment monitoring period; for line units with high misalignment sensitivity coefficients SM, the period is shortened so that misalignment analysis can be performed multiple times, thereby improving the accuracy of line misalignment abnormality analysis; for line units with low misalignment sensitivity coefficients SM, the period is extended so as to reduce the number of misalignment analyses and reduce costs, thereby improving the efficiency of misalignment monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0039] Figure 1 This is a system block diagram of a medium and high voltage misalignment monitoring system based on big data analysis in the present invention.

[0040] Figure 2It is a connection diagram of the data acquisition module in the present invention.

[0041] Figure 3 This is a training flow chart of the misalignment phenomenon diagnosis model in the present invention. DETAILED DESCRIPTION

[0042] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figures 1 to 3 , the first aspect of the present invention provides a medium and high voltage misalignment monitoring system based on big data analysis, including a clock synchronization module, a data acquisition module, an industrial server, a misalignment analysis module, an operation and maintenance management module, a model building module, a cloud platform and a collaborative evaluation module;

[0044] The clock synchronization module is connected to the data acquisition module, and gives precise timing tags to the data collected by the data acquisition module, effectively realizing synchronous transmission of data, solving the problem of network delay caused by remote data transmission, which leads to the inability to synchronize real-time data with displayed data.

[0045] The data collection module includes three switches, which are used to collect power data of each line unit in the monitoring area according to a preset cycle; the power data includes power consumption at the power supply metering point, power consumption at the load metering point, and fixed power loss; the specific collection steps are as follows:

[0046] Based on the power supply mode and line unit of medium and high voltage lines, the metering points on the medium and high voltage feeders are assigned to the line units; the metering points on the medium and high voltage feeders include the metering points on the outlet side of the substation, the metering points corresponding to the tie switches between adjacent lines, and the metering points corresponding to all transformers on the corresponding lines;

[0047] It can be understood that the line unit of the present application includes a power supply metering point and a plurality of load metering points; the relationship between the power supply metering point and the load metering point forms a simple tree topology, and the line power supply metering point provides energy input for the next level load metering point;

[0048] The three switches are all connected to the power consumption information collection system, and one of the switches is used to collect the power consumption of the power supply metering point within the week of the day detected by the power consumption information collection system;

[0049] One switch is used to collect the power consumption of each load metering point within a week of the day detected by the power consumption information collection system; one switch is used to collect the fixed power loss within a week of the day detected by the power consumption information collection system; the fixed power loss includes line loss and metering equipment loss;

[0050] In this application, the power consumption of each metering point on the medium and high voltage feeder is collected through the switch. Under the action of the clock synchronization module, the power consumption of metering points at different locations can be synchronously collected and analyzed, so as to realize the inaccuracy monitoring of the whole network and assist in determining the fault location with the help of its own position information when a fault occurs, shorten the time of fault investigation, and improve the reliability of the distribution network;

[0051] The industrial server is used to aggregate and process the data information collected by the switch and transmit the power data of each line unit to the misalignment analysis module; the power data carries a unique line identifier;

[0052] The misalignment analysis module is used to input the power data of each line unit into the misalignment phenomenon diagnosis model and obtain the output result to determine whether there is a misalignment phenomenon;

[0053] When the output result is 1, it is determined that the corresponding line unit has misalignment, and a misalignment signal is generated to the industrial server; when the output result is 0, it is determined that the corresponding line unit is measuring normally;

[0054] When the line unit is misaligned, the industrial server sends a misalignment signal to the operation and maintenance management module. After receiving the misalignment signal, the operation and maintenance management module arranges power personnel to check the load metering points in the line unit one by one to obtain the misalignment data of the load metering points. Of course, it can also focus on checking the daily working status of the load metering points.

[0055] In a possible implementation, the model building module is used to periodically acquire historical power data of each line unit to build a sample data set to train a misalignment phenomenon diagnosis model;

[0056] The specific training steps of the model building module are as follows:

[0057] S1: sorting and extracting sample data of a given time period to form input and output data sets; the input data refers to the normal power data and inaccurate power data of the line unit within a certain time period;

[0058] The output data is whether there is a misalignment phenomenon, which is represented by 0 and 1; 0 means that the line unit measurement is normal, and 1 means that the line unit has a misalignment phenomenon;

[0059] S2: Divide the dataset into a training set, a validation set, and a test set according to preset ratios; the preset ratios include 2:1:1, 3:1:1, and 4:3:1;

[0060] S3: Constructing a fusion model: The fusion model is a model constructed by combining at least two of the support vector machine, deep convolutional neural network and RBF neural network. The fusion methods include linear weighted fusion, cross fusion, waterfall fusion, feature fusion and prediction fusion.

[0061] S4: After normalizing the data of the training set, the test set and the verification set, the fusion model is trained, tested and verified, and the trained fusion model is marked as the misalignment diagnosis model;

[0062] The present application inputs the power data of the line unit into the misalignment phenomenon diagnosis model to obtain the output result to determine whether there is a misalignment phenomenon. The misalignment data can be analyzed in a full range to improve the comprehensiveness of misalignment monitoring. At the same time, fewer computing resources are required, making the monitoring process more efficient and reducing costs.

[0063] In a possible implementation, since the data acquisition module collects the power data of each line unit in the monitoring area according to a preset period, and then summarizes and analyzes it;

[0064] It is understandable that the conditions such as metering equipment, operating environment, and device reliability involved in each line unit are different, so the misalignment monitoring period of each line unit is different;

[0065] The collaborative evaluation module is used to evaluate the misalignment sensitivity coefficient SM of each line unit, so as to adjust the misalignment monitoring period according to the misalignment sensitivity coefficient SM. The specific evaluation process is as follows:

[0066] For a certain line unit, the number of load metering points of the line unit is counted as L1; the number of user sides accessing the line unit is counted as H1;

[0067] In the preset time period, the average daily power consumption of the line unit is calculated as D1; ​​it can be understood that the average daily power consumption of the line unit refers to the sum of the average daily power consumption of the user side connected to the line unit;

[0068] Collect misalignment event records of the line unit; count the total number of misalignments of the line unit as C1; use the formula SM = (H1×a1+D1×a2+C1×a3) / (L1×a4) to calculate the misalignment sensitivity coefficient SM of the line unit, where a1, a2, a3, and a4 are all preset coefficient factors;

[0069] For example: In the past three months, the number of load metering points of the line unit is 10, the number of users is 100, the average daily power consumption is 500 degrees, and the total number of misalignment is 5 times;

[0070] Remove the dimension of each parameter and calculate its value; a1 takes the value of 0.01, a2 takes the value of 0.1, a3 takes the value of 0.5, and a4 takes the value of 0.2; then the misalignment sensitivity coefficient SM = (100×0.01+500×0.1+5×0.5) / (10×0.2) = 26.75;

[0071] The misalignment monitoring period of the line unit is adaptively adjusted according to the misalignment sensitivity coefficient SM, for example, the monitoring and analysis is adjusted from once every 10 days to once every 7 days; specifically:

[0072] A comparison table of misalignment sensitivity coefficient range and misalignment monitoring period is pre-stored in the cloud platform;

[0073] According to the comparison table, determine the misalignment sensitivity coefficient range that matches the misalignment sensitivity coefficient SM, and then determine the corresponding misalignment monitoring period according to the misalignment sensitivity coefficient range; the larger the misalignment sensitivity coefficient SM is, the shorter the corresponding misalignment monitoring period is;

[0074] The collaborative evaluation module is used to feed back the adjusted misalignment monitoring cycle to the industrial server; the industrial server is used to generate a data collection instruction to the switch according to the misalignment monitoring cycle;

[0075] The present application can reasonably adjust the misalignment monitoring period of the line unit according to the misalignment sensitivity coefficient SM; for the line unit with a high misalignment sensitivity coefficient SM, the period is shortened so as to perform misalignment analysis multiple times, thereby improving the accuracy of line misalignment abnormality analysis; for the line unit with a low misalignment sensitivity coefficient SM, the period is extended so as to reduce the number of misalignment analyses and reduce costs, thereby improving the misalignment monitoring efficiency.

[0076] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0077] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0078] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A medium and high voltage misalignment monitoring system based on big data analysis, characterized in that: It includes clock synchronization module, data acquisition module, industrial server, misalignment analysis module, operation and maintenance management module, model building module, cloud platform and collaborative evaluation module; The data acquisition module includes three switches for collecting power data of each line unit in the monitoring area according to a preset period; the power data includes power consumption at the power supply metering point, power consumption at the load metering point, and fixed power loss; The industrial server is used to aggregate and process the data information collected by the switch, and transmit the power data of each line unit to the misalignment analysis module; the power data carries a unique line identifier; The model building module is used to periodically acquire historical power data of each line unit to build a sample data set to train a misalignment phenomenon diagnosis model; The misalignment analysis module is used to input the power data of each line unit into the misalignment phenomenon diagnosis model, obtain the output result, and determine whether there is a misalignment phenomenon; When the output result is 1, it is determined that the corresponding line unit has misalignment, and a misalignment signal is generated to the industrial server; When the output result is 0, it is determined that the corresponding line unit measurement is normal; When the line unit is misaligned, the industrial server sends a misalignment signal to the operation and maintenance management module; after receiving the misalignment signal, the operation and maintenance management module arranges power personnel to check the load metering points in the line unit one by one to obtain the misalignment data of the load metering points; The collaborative evaluation module is used to evaluate the misalignment sensitivity coefficient SM in combination with the metering points of the line unit and the user side distribution, so as to adaptively adjust the corresponding misalignment monitoring period according to the misalignment sensitivity coefficient SM; the specific evaluation process is as follows: For a certain line unit, the number of load metering points of the line unit is counted as L1; the number of user sides accessing the line unit is counted as H1; In a preset time period, the average daily power consumption of the line unit is counted as D1; ​​the misalignment event records of the line unit are collected, and the total number of misalignment times of the line unit is counted as C1; The misalignment sensitivity coefficient SM of the line unit is calculated using the formula SM=(H1×a1+D1×a2+C1×a3) / (L1×a4), where a1, a2, a3, and a4 are all preset coefficient factors; The corresponding misalignment monitoring period is adaptively adjusted according to the misalignment sensitivity coefficient SM, specifically: A comparison table of misalignment sensitivity coefficient range and misalignment monitoring period is pre-stored in the cloud platform; According to the comparison table, determining a misalignment sensitivity coefficient range that matches the misalignment sensitivity coefficient SM, and then determining a corresponding misalignment monitoring period according to the misalignment sensitivity coefficient range; The collaborative evaluation module is used to feed back the adjusted misalignment monitoring cycle to the industrial server; the industrial server is used to generate a data collection instruction to the switch according to the misalignment monitoring cycle.

2. A medium and high voltage misalignment monitoring system based on big data analysis according to claim 1, characterized in that: The specific collection steps of the data collection module are as follows: Based on the medium and high voltage line power supply mode and line unit, the metering points on the medium and high voltage feeder are assigned to the line unit; the line unit includes a power supply metering point and several load metering points; The three switches are all connected to the power consumption information collection system, and one of the switches is used to collect the power consumption of the power supply metering point within the week of the day detected by the power consumption information collection system; One switch is used to collect the power consumption of each load metering point within the week of the day detected by the power consumption information collection system; one switch is used to collect the fixed power supply loss within the week of the day detected by the power consumption information collection system; the fixed power supply loss includes line loss and metering equipment loss.

3. A medium and high voltage misalignment monitoring system based on big data analysis according to claim 2, characterized in that: The relationship between the power supply metering point and the load metering point forms a simple tree topology, and the line power supply metering point provides energy input for the next level load metering point.

4. A medium and high voltage misalignment monitoring system based on big data analysis according to claim 2, characterized in that: The metering points on medium and high voltage feeders include the metering points on the outlet side of the substation, the corresponding metering points of the interconnecting switches between adjacent lines, and the corresponding metering points of all transformers on the corresponding lines.

5. The medium and high voltage misalignment monitoring system based on big data analysis according to claim 1 is characterized in that: The clock synchronization module is connected to the data acquisition module, and gives precise timing tags to the data collected by the data acquisition module to achieve synchronous transmission of the data.

6. The medium and high voltage misalignment monitoring system based on big data analysis according to claim 1 is characterized in that: The specific training steps of the model building module are as follows: S1: sorting and extracting sample data of a given time period to form input and output data sets; the input data refers to the normal power data and inaccurate power data of the line unit within a certain time period; The output data is whether there is a misalignment phenomenon, which is represented by 0 and 1; 0 means that the line unit measurement is normal, and 1 means that the line unit has a misalignment phenomenon; S2: Divide the data set into a training set, a validation set, and a test set according to a preset ratio; the preset ratio includes 2:1:1, 3:1:1, and 4:3:1; S3: Constructing a fusion model: The fusion model is a model constructed by combining at least two of the support vector machine, deep convolutional neural network and RBF neural network; S4: After normalizing the data of the training set, test set and verification set, the fusion model is trained, tested and verified, and the trained fusion model is marked as the misalignment diagnosis model.

7. A medium and high voltage misalignment monitoring system based on big data analysis according to claim 6, characterized in that: Fusion methods include linear weighted fusion, cross fusion, waterfall fusion, feature fusion and predictive fusion.

8. The medium and high voltage misalignment monitoring system based on big data analysis according to claim 1 is characterized in that: The operation and maintenance management module is used to arrange power personnel to conduct key inspections based on the daily working status of the load metering points to obtain the inaccurate data of the load metering points.