CN118311352A - Bus duct fault diagnosis method and system for photovoltaic energy storage system - Google Patents

Abstract

The present disclosure provides a bus duct fault diagnosis method and system for a photovoltaic energy storage system, which relates to the technical field of equipment fault diagnosis, and the method includes: performing multiple warning tests on the bus duct under multiple current data according to multi-dimensional warning indicators to build a current-warning threshold database; inputting the predicted current into the current-warning threshold database to match and obtain a multi-dimensional fault warning threshold; judging the sensor monitoring data of the bus duct based on the multi-dimensional fault warning threshold; marking the abnormal slot number according to the judgment result; and performing bus duct fault inspection and maintenance based on the abnormal slot number. The present disclosure can solve the technical problem in the prior art that the accuracy of the warning judgment of the bus duct is low due to the inability to set an accurate fault warning threshold in combination with the actual situation of photovoltaic power generation, resulting in the inability to timely and effectively carry out the bus duct fault warning, and can improve the accuracy and timeliness of the fault warning, so as to timely and effectively carry out the fault abnormality warning.

Description

Bus duct fault diagnosis method and system for photovoltaic energy storage system

Technical Field

The present disclosure relates to the technical field of equipment fault diagnosis, and in particular to a bus duct fault diagnosis method and system for a photovoltaic energy storage system.

Background technique

The bus duct of a photovoltaic energy storage system is a device used to connect and transmit electrical energy between photovoltaic panels and energy storage equipment. Its main function is to collect the DC power generated by photovoltaic panels and transmit it to the energy storage equipment for storage, or to invert it into AC power and supply it to the load.

Since photovoltaic power generation is greatly affected by environmental factors such as light and weather, the power generation current is highly volatile. When the current in the bus duct increases, the temperature rise of the bus duct will increase, and it will also cause the bus duct to vibrate and generate noise. These conditions are normal when the current flowing through is large; but when the current flowing through is small, these conditions may be a precursor to bus duct failure. Therefore, the existing use of fixed warning thresholds in dimensions such as temperature, vibration, and noise for fault warning will make the warning threshold less compatible with the actual situation of photovoltaic power generation, resulting in inaccurate setting of the warning threshold and low accuracy of the bus duct warning judgment.

In summary, the existing bus duct fault diagnosis method is unable to set an accurate fault warning threshold in combination with the actual situation of photovoltaic power generation, resulting in low accuracy of bus duct warning judgment, causing the technical problem of being unable to provide timely and effective bus duct fault warning.

Summary of the invention

The purpose of the present invention is to provide a bus duct fault diagnosis method and system for a photovoltaic energy storage system, so as to solve the technical problem that the existing bus duct fault diagnosis method cannot set an accurate fault warning threshold in combination with the actual situation of photovoltaic power generation, resulting in low accuracy of bus duct warning judgment and the inability to provide bus duct fault warning in a timely and effective manner.

In view of the above problems, the present disclosure provides a bus duct fault diagnosis method and system for a photovoltaic energy storage system.

In a first aspect, the present disclosure provides a bus duct fault diagnosis method for a photovoltaic energy storage system, the method being implemented by a bus duct fault diagnosis system for a photovoltaic energy storage system, wherein the method comprises: transmitting the light intensity and real-time temperature collected within a preset time window to a current prediction channel for current prediction, and outputting a predicted current, wherein the current prediction channel is constructed based on a target photovoltaic device; obtaining an expected current threshold, performing multiple warning tests on the bus duct under multiple current data within the expected current threshold according to multi-dimensional warning indicators, and constructing a current-warning threshold database according to the test results; inputting the predicted current into the current-warning threshold database, and matching to obtain a fault warning threshold, wherein the fault warning threshold includes a temperature warning threshold, a vibration warning threshold, and a noise warning threshold; receiving a sensor monitoring data set for the bus duct, wherein the sensor monitoring data includes a temperature monitoring data set; The sensor monitoring data is collected from the sensor monitoring data set, and each slot of the bus duct corresponds to a sensor monitoring data, wherein the sensor monitoring data is marked with a slot number; the temperature warning threshold is corrected according to the real-time temperature to obtain an updated temperature warning threshold, and the sensor monitoring data in the sensor monitoring data set are judged based on the updated temperature warning threshold, the vibration warning threshold and the noise warning threshold; when the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, the corresponding slot number is extracted and abnormally marked to obtain an abnormal slot number set; positioning and identification are performed based on the abnormal slot number set, and the position coordinates of the abnormal slot are sent to the nearest fault inspection and maintenance personnel to perform bus duct fault inspection and maintenance.

In a second aspect, the present disclosure further provides a bus duct fault diagnosis system for a photovoltaic energy storage system, which is used to execute a bus duct fault diagnosis method for a photovoltaic energy storage system as described in the first aspect, wherein the system includes: a current prediction module, the current prediction module is used to transmit the light intensity and real-time temperature collected within a preset time window to a current prediction channel for current prediction, and output a predicted current, the current prediction channel is constructed based on a target photovoltaic device; a current-warning threshold database construction module, the current-warning threshold database construction module is used to obtain an expected current threshold, perform multiple warning tests on the bus duct under multiple current data within the expected current threshold according to multi-dimensional warning indicators, and construct a current-warning threshold database according to the test results; a fault warning threshold matching module, the fault warning threshold matching module is used to input the predicted current into the current-warning threshold database, match to obtain a fault warning threshold, the fault warning threshold includes a temperature warning threshold, a vibration warning threshold and a noise warning threshold; a sensor monitoring data set receiving module, the sensor monitoring data set receiving module is used to receive a sensor monitoring data set of the bus duct, The sensor monitoring data includes temperature monitoring data, vibration monitoring data and noise monitoring data, and each slot of the bus duct corresponds to a sensor monitoring data, wherein the sensor monitoring data is marked with a slot number; a sensor monitoring data judgment module, the sensor monitoring data judgment module is used to correct the temperature warning threshold according to the real-time temperature, obtain an updated temperature warning threshold, and judge the sensor monitoring data in the sensor monitoring data set based on the updated temperature warning threshold, vibration warning threshold and noise warning threshold; an abnormal slot number set acquisition module, the abnormal slot number set acquisition module is used to extract and abnormally mark the corresponding slot number when the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, so as to obtain an abnormal slot number set; a bus duct fault inspection and maintenance module, the bus duct fault inspection and maintenance module is used to locate and identify based on the abnormal slot number set, send the abnormal slot position coordinates to the nearest fault inspection and maintenance personnel, and perform bus duct fault inspection and maintenance.

One or more technical solutions provided in this disclosure have at least the following technical effects or advantages:

1. Current prediction is performed by transmitting the light intensity and real-time temperature collected within a preset time window to a current prediction channel, and a predicted current is output. The current prediction channel is constructed based on a target photovoltaic device; an expected current threshold is obtained, and multiple warning tests are performed on the bus duct under multiple current data within the expected current threshold according to multi-dimensional warning indicators, and a current-warning threshold database is constructed according to the test results; the predicted current is input into the current-warning threshold database, and a fault warning threshold is obtained by matching, and the fault warning threshold includes a temperature warning threshold, a vibration warning threshold, and a noise warning threshold; a sensor monitoring data set of the bus duct is received, wherein the sensor monitoring data includes temperature monitoring data, vibration monitoring data, and noise monitoring data, and each slot of the bus duct corresponds to a sensor Monitoring data, wherein the sensor monitoring data is marked with a slot number; the temperature warning threshold is corrected according to the real-time temperature to obtain an updated temperature warning threshold, and the sensor monitoring data in the sensor monitoring data set are judged based on the updated temperature warning threshold, the vibration warning threshold and the noise warning threshold; when the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, the corresponding slot number is extracted and abnormally marked to obtain an abnormal slot number set; positioning and identification are performed based on the abnormal slot number set, and the abnormal slot position coordinates are sent to the nearest fault inspection and maintenance personnel to perform bus duct fault inspection and maintenance. In other words, the above method can solve the technical problem that the existing bus duct fault diagnosis method cannot set an accurate fault warning threshold in combination with the actual situation of photovoltaic power generation, resulting in low accuracy of bus duct warning judgment, which causes the inability to timely and effectively perform bus duct fault warning.

2. By dynamically adjusting the fault warning threshold based on photovoltaic power generation prediction, the matching degree between the fault warning threshold and the actual power generation of photovoltaic equipment can be improved, thereby improving the accuracy of the fault warning threshold setting.

3. By combining the fault warning thresholds of multiple dimensions to make early warning judgments on the real-time operating status of the bus duct, the accuracy and timeliness of the bus duct fault warning can be improved, so that fault abnormality warnings can be carried out in a timely and effective manner to avoid major safety losses.

The above description is only an overview of the technical solution of the present disclosure. In order to more clearly understand the technical means of the present disclosure, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific implementation methods of the present disclosure are listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easy to understand through the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the present disclosure or 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 in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

FIG1 is a schematic diagram of a process of diagnosing bus duct faults in a photovoltaic energy storage system according to the present disclosure;

FIG2 is a schematic diagram of a process for constructing a current prediction channel in a bus duct fault diagnosis method for a photovoltaic energy storage system disclosed in the present invention;

FIG3 is a schematic diagram of the structure of a bus duct fault diagnosis system for a photovoltaic energy storage system disclosed in the present invention.

Description of reference numerals:

Current prediction module 11, current-warning threshold database construction module 12, fault warning threshold matching module 13, sensor monitoring data set receiving module 14, sensor monitoring data judgment module 15, abnormal slot number set obtaining module 16, bus duct fault inspection and maintenance module 17.

Detailed ways

The present disclosure provides a bus duct fault diagnosis method and system for a photovoltaic energy storage system, thereby solving the technical problem that the existing bus duct fault diagnosis method cannot set an accurate fault warning threshold value in combination with the actual situation of photovoltaic power generation, resulting in low accuracy of early warning judgment of the bus duct and failure to promptly and effectively provide early warning of bus duct faults. The present disclosure achieves the technical effect of improving the accuracy and timeliness of bus duct fault warnings, thereby enabling timely and effective abnormal early warning of bus duct faults.

Below, the technical solutions in the present disclosure will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments of the present disclosure. It should be understood that the present disclosure is not limited to the example embodiments described herein. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure. It should also be noted that, for the convenience of description, only the parts related to the present disclosure are shown in the accompanying drawings, rather than all of them.

Embodiment 1

Please refer to FIG. 1 . The present disclosure provides a bus duct fault diagnosis method for a photovoltaic energy storage system. The method is applied to a bus duct fault diagnosis system for a photovoltaic energy storage system. The method specifically includes the following steps:

Step 1: transmitting the light intensity and real-time temperature collected within a preset time window to a current prediction channel for current prediction, and outputting a predicted current, wherein the current prediction channel is constructed based on a target photovoltaic device;

Specifically, first, a preset time window is obtained, which is a short period of time, and can be set by a person skilled in the art according to actual conditions, such as 30 minutes, 60 minutes, etc. Then, within the preset time window, the current light intensity and real-time ambient temperature of the photovoltaic device are collected through multiple sensors to obtain light intensity data and real-time temperature data.

Then the light intensity data and real-time temperature data are input into a pre-built current prediction channel for current prediction, and a current prediction result, i.e., the predicted current, is obtained. The current prediction channel is built based on the target photovoltaic device, and is a feedforward neural network model that can be iteratively optimized in machine learning, and is obtained through supervised training of historical data sets. By obtaining the predicted current, support is provided for the next step of fault warning data matching.

Step 2: Obtain the expected current threshold, perform multiple early warning tests on the bus duct under multiple current data within the expected current threshold according to the multi-dimensional early warning indicators, and build a current-early warning threshold database according to the test results;

Specifically, first, the expected current threshold is obtained, and the expected current threshold refers to the power generation current range of the photovoltaic device during the power transmission process, which can be constructed by extracting the minimum current value and the maximum current value in the historical power generation data of the photovoltaic device. Then, according to the multi-dimensional warning index, multiple warning tests are performed on the bus duct under multiple current data within the expected current threshold. When the current in the bus duct increases, the temperature rise of the bus duct will increase, and it will also cause the bus duct to vibrate and generate noise. Therefore, the multi-dimensional warning index is set as a temperature index, a vibration index and a noise index, and multiple warning test results are obtained. A current-warning threshold database is constructed according to the multiple warning test results, wherein the current-warning threshold database stores multiple current values and multiple warning threshold sets, and each current value corresponds to a warning threshold set.

By conducting multiple warning tests to build a current-warning threshold database, the accuracy of the current-warning threshold database setting can be improved. At the same time, it provides support for the next step of warning threshold matching under actual current conditions, which can improve the efficiency and accuracy of warning threshold matching.

Step 3: inputting the predicted current into the current-warning threshold database, matching to obtain a fault warning threshold, wherein the fault warning threshold includes a temperature warning threshold, a vibration warning threshold and a noise warning threshold;

Specifically, the predicted current is input into the current-warning threshold database for matching, and the fault warning threshold corresponding to the predicted current is obtained, wherein the fault warning threshold includes a temperature warning threshold, a vibration warning threshold and a noise warning threshold, wherein the temperature warning threshold is the warning temperature; the vibration warning threshold is one or more abnormal vibration signals, including abnormal vibration frequency and abnormal amplitude; the noise warning threshold is one or more abnormal noise signals, including abnormal noise frequency, abnormal noise tone and abnormal noise intensity. By obtaining the fault warning threshold, a basis is provided for the next step of real-time status warning judgment of the bus duct.

Step 4: receiving a sensor monitoring data set of the bus duct, wherein the sensor monitoring data includes temperature monitoring data, vibration monitoring data and noise monitoring data, and each slot of the bus duct corresponds to a sensor monitoring data, wherein the sensor monitoring data is marked with a slot number;

Specifically, the bus duct is composed of multiple trough units, each trough unit has a corresponding unique number, and each trough unit is configured with a temperature sensor, a vibration sensor and a noise sensor, wherein the number of sensors is one or more and can be set based on the actual length and area of the trough unit. If there are multiple sensors, the sensor monitoring data is the average of the monitoring data of multiple sensors.

Multiple monitoring data collection nodes are set within the preset time window, wherein the monitoring data collection nodes can be set according to actual needs, wherein the higher the quality of the demand warning, the shorter the time interval of the monitoring data collection node, for example: assuming that the preset time window is 30 minutes, the time interval of the monitoring data collection node can be set to 1 minute. The sensor monitoring data of the bus duct is received at multiple monitoring data collection nodes to obtain a sensor monitoring data set, wherein the sensor monitoring data includes temperature monitoring data, vibration monitoring data and noise monitoring data, wherein each slot of the bus duct corresponds to a sensor monitoring data, and the sensor monitoring data is marked with a slot number.

By acquiring the real-time sensor monitoring data of the bus duct, the real-time operating status information of the bus duct can be intuitively obtained, and at the same time, data support is provided for the fault warning judgment of the bus duct.

Step 5: Correcting the temperature warning threshold according to the real-time temperature to obtain an updated temperature warning threshold, and judging the sensor monitoring data in the sensor monitoring data set based on the updated temperature warning threshold, the vibration warning threshold and the noise warning threshold;

Specifically, since the temperature warning threshold obtained by matching is obtained by analysis at a fixed ambient temperature, it is necessary to calibrate the temperature warning threshold according to the real-time temperature to further improve the accuracy of the temperature warning threshold setting and obtain an updated temperature warning threshold. Then, the temperature monitoring data in the sensor monitoring data is judged according to the updated temperature warning threshold, the vibration monitoring data in the sensor monitoring data is judged according to the vibration warning threshold, and the noise monitoring data in the sensor monitoring data is judged according to the noise warning threshold.

Step 6: When the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, the corresponding slot number is extracted and abnormally marked to obtain an abnormal slot number set;

Specifically, when the temperature monitoring data is greater than the updated temperature warning threshold and/or the vibration monitoring data satisfies the vibration warning threshold and/or the noise monitoring data satisfies the noise warning threshold, wherein the vibration monitoring data satisfies the vibration warning threshold means that the vibration monitoring data meets any one of the abnormal vibration signals in the vibration warning threshold; the noise monitoring data satisfies the noise warning threshold means that the noise monitoring data meets any one of the abnormal noise signals in the noise warning threshold, then the slot number corresponding to the sensor monitoring data is extracted, and the slot number is marked as abnormal to obtain multiple abnormal slot numbers, and an abnormal slot number set is formed based on the multiple abnormal slot numbers.

By obtaining the abnormal slot number set, it provides support for the accurate positioning of the abnormal slot, and at the same time can improve the efficiency and accuracy of abnormal slot fault inspection and repair.

Step 7: Perform positioning and identification based on the abnormal slot number set, send the abnormal slot position coordinates to the nearest fault inspection and maintenance personnel, and perform bus duct fault inspection and maintenance.

Specifically, the abnormal slot is accurately located according to the abnormal slot number set to determine the position coordinates of the abnormal slot. The position coordinates of multiple fault maintenance personnel currently on duty are obtained, and the abnormal slot maintenance task is assigned based on the position coordinates of multiple fault maintenance personnel, and the abnormal slot position coordinates are sent to the nearest fault maintenance personnel, and finally the fault maintenance personnel perform bus duct fault maintenance and maintenance according to the received abnormal slot position coordinates.

Specifically, the bus duct fault diagnosis method for a photovoltaic energy storage system is applied to a bus duct fault diagnosis system for a photovoltaic energy storage system.

First, the light intensity and real-time temperature collected within a preset time window are transmitted to a current prediction channel for current prediction, and a predicted current is output. The current prediction channel is constructed based on a target photovoltaic device. Then, an expected current threshold is obtained, and multiple warning tests are performed on the bus duct under multiple current data within the expected current threshold according to multi-dimensional warning indicators, and a current-warning threshold database is constructed according to the test results. The predicted current is further input into the current-warning threshold database, and a fault warning threshold is obtained by matching. The fault warning threshold includes a temperature warning threshold, a vibration warning threshold, and a noise warning threshold. On the other hand, a sensor monitoring data set of the bus duct is received, wherein the sensor monitoring data includes temperature monitoring data, vibration monitoring data, and noise monitoring data, and each slot of the bus duct corresponds to a sensor monitoring data, wherein the sensor monitoring data is marked with a slot number; the temperature warning threshold is corrected according to the real-time temperature to obtain an updated temperature warning threshold, and the sensor monitoring data in the sensor monitoring data set are judged based on the updated temperature warning threshold, the vibration warning threshold and the noise warning threshold; when the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, the corresponding slot number is extracted and abnormally marked to obtain an abnormal slot number set; finally, positioning and identification are performed based on the abnormal slot number set, and the abnormal slot position coordinates are sent to the nearest fault inspection and maintenance personnel to perform bus duct fault inspection and maintenance.

By dynamically adjusting the fault warning threshold based on photovoltaic power generation prediction, the accuracy of the fault warning threshold setting can be improved. At the same time, the real-time operation status of the bus duct can be warned by combining the fault warning thresholds of multiple dimensions, which can improve the accuracy and timeliness of the bus duct fault warning, so that the abnormal bus duct fault warning can be carried out in a timely and effective manner to avoid major safety losses.

Further, as shown in FIG. 2 , step 1 of the present disclosure includes:

Obtaining basic indicator data of target photovoltaic equipment, wherein the basic indicator data includes equipment type, installed capacity, and photoelectric conversion efficiency;

Based on the industrial big data, the photovoltaic power generation related data is retrieved with the basic indicator data as the retrieval condition to obtain a plurality of sample power generation data, wherein the sample power generation data includes sample light intensity, sample temperature and sample power generation current;

performing data cleaning on the plurality of sample power generation data to obtain a plurality of standard sample power generation data;

The plurality of standard sample power generation data are used as training data, and supervised learning is performed on the current prediction channel constructed based on the BP neural network to obtain a current prediction channel that meets expected indicators.

Specifically, first, basic indicator data of the target photovoltaic equipment is obtained, wherein the basic indicator data includes equipment type, installed capacity and photoelectric conversion efficiency, and those skilled in the art may set the basic indicator data according to the actual situation of the target photovoltaic equipment.

Industrial big data is a series of technologies and methods that enable the value contained in massive industrial data to be mined and displayed, including data planning, data collection, analysis and mining, etc. Based on industrial big data technology, the basic indicator data of the target photovoltaic equipment is used as the search condition to retrieve photovoltaic power generation related data, and multiple sample power generation data are obtained, wherein the sample power generation data includes sample light intensity, sample temperature and sample power generation current.

The plurality of sample power generation data are cleaned according to a preset data cleaning scheme, and the preset data cleaning scheme at least includes steps such as data deduplication, missing value supplementation, and outlier processing, which are commonly used data processing methods for those skilled in the art and will not be described in detail here, to obtain a plurality of standard sample power generation data after data cleaning. By performing data cleaning on the sample power generation data, the accuracy of obtaining the standard sample power generation data can be improved, thereby indirectly improving the accuracy of the current prediction channel training.

A current prediction channel is constructed based on a BP neural network. The current prediction channel is a feedforward neural network model that can be iteratively optimized in machine learning. It is obtained through supervised training of historical data sets. The current prediction channel includes an input layer, a hidden layer, and an output layer. The input data of the input layer are light intensity and temperature, and the output data is the power generation current. Then, the multiple standard sample power generation data are used as training data to perform supervised training on the current prediction channel.

The method for supervised training of the current prediction channel is as follows: first, the training data is divided into a sample training set and a sample verification set according to a preset data division ratio. The preset data division ratio can be set according to the actual sample data volume. Normally, the sample training set accounts for 80% and the sample verification set accounts for 20%. Then, supervised training is performed on the current prediction channel through the sample training set. First, first sample training data is randomly selected from the sample training set, and supervised training is performed on the current prediction channel through the first sample training data to obtain a first power generation current; the first power generation current is compared with the first sample power generation current in the first sample training data; when the two are consistent, the second sample training data is randomly selected to supervise the current prediction channel; when the two are inconsistent, the deviation value between the first power generation current and the first sample power generation current in the first sample training data is calculated, and the weight parameter of the current prediction channel is optimized and adjusted according to the deviation value, and then the second sample training data is randomly selected to supervise the current prediction channel; iterative training is continuously performed through the sample data set. When the output result of the current prediction channel tends to converge, the current prediction channel is then verified and trained through the sample verification set until the accuracy of the output result of the current prediction channel meets the expected index, and the trained current prediction channel is obtained. The expected index is the output accuracy index, which can be set according to actual needs. The higher the required accuracy, the greater the output accuracy index. For example, the output accuracy index can be set to an accuracy of 95%.

By constructing a current prediction channel based on a BP neural network and obtaining corresponding sample power generation training data to perform supervised training on the current prediction channel, support is provided for power generation current prediction and the accuracy of power generation current prediction can be improved.

Further, step 2 of the present disclosure includes:

The multi-dimensional early warning indicators include temperature indicators, vibration indicators and noise indicators;

Based on a preset analysis step length, the expected current threshold is divided into equal values to obtain a plurality of current data;

Constructing a bus duct twin model, and sequentially performing multiple early warning tests under multiple current data through the bus duct twin model to obtain multiple early warning indicator sets;

Specifically, the method for constructing a current-warning threshold database is as follows: first, a multi-dimensional warning indicator is obtained, wherein the multi-dimensional warning indicator includes a temperature indicator, a vibration indicator, and a noise indicator. Then, the expected current threshold is divided into equal values based on a preset analysis step to obtain a plurality of current data, wherein the preset analysis compensation can be set according to the range of the expected current threshold and the actual accuracy requirement, wherein the smaller the preset analysis step, the higher the analysis accuracy, for example: assuming that the expected current threshold is 100 to 200 amps, the preset analysis step can be set to 0.5 amps, that is, a current data is set every 0.5 amps, such as: 100.5 amps, 101 amps, etc.

A bus duct twin model is constructed based on the digital twin technology, and then multiple early warning tests under multiple current data are performed in sequence through the bus duct twin model, and multiple early warning indicator sets are obtained according to the test results.

Furthermore, the present disclosure also includes the following steps:

Obtaining equipment specification data, operation control parameters and working environment data of the bus duct;

In the visual simulation platform, simulation modeling of the bus duct equipment is performed based on the equipment specification data and the operation control parameters to obtain a twin model of the bus duct equipment;

The bus duct equipment twin model is environmentally configured based on the working environment data to obtain the bus duct twin model.

Specifically, the method for constructing a bus duct twin model is as follows: first, the equipment specification data, operation control parameters and working environment data of the bus duct are obtained. The equipment specification data includes information such as equipment type and metal bus size. The operation control parameters include operating voltage, operating current and other data. The working environment data refers to the environmental parameters of the bus duct working area. The environmental parameter settings with the highest frequency of occurrence can be selected, including ambient temperature, ambient humidity and other data.

Digital twin technology is a method of creating a highly simulated virtual model for a real object in a digital way to achieve a virtual representation of the state of a physical entity or system, with multiple advantages such as real-time, fidelity, interoperability, and closed-loop. Based on digital twin technology, in a visual simulation platform, the bus duct equipment is simulated and modeled according to the equipment specification data and the operation control parameters. Commonly used visual simulation platforms include Blender platform, AutoCAD platform, etc., and an adaptive visual simulation platform can be selected according to actual conditions for simulation modeling to obtain a bus duct equipment twin model.

The bus duct equipment twin model is configured according to the working environment data to obtain a bus duct twin model. By constructing a bus duct twin model based on digital twin technology, the authenticity and accuracy of the bus duct simulation test can be improved, thereby improving the accuracy and rationality of the early warning indicators.

Furthermore, the present disclosure also includes the following steps:

randomly selecting first current data from the plurality of current data;

Inputting the first current data into the bus duct twin model, performing multiple fault warning tests under a preset number threshold, and obtaining multiple first initial warning indicator sets, wherein the first initial warning indicator set includes a first temperature, a first vibration characteristic data, and a first noise characteristic data;

Performing data integration on the multiple first initial early warning indicator sets to obtain multiple first early warning indicator sets;

The first warning indicator with the highest occurrence frequency in the multiple first warning indicator sets is extracted to construct a first warning indicator set, and the first warning indicator set is added to the multiple warning indicator sets.

Specifically, the method for obtaining multiple early warning indicator sets is as follows: first, randomly select the first current data from the multiple current data, and the first current data is any one of the multiple current data. Then the first current data is input into the bus duct twin model, and multiple fault early warning tests under a preset number threshold are performed. The preset number threshold can be set according to actual conditions, wherein the larger the preset number threshold, the higher the accuracy of the early warning indicator, and multiple first initial early warning indicator sets are obtained, wherein the first initial early warning indicator set includes the first temperature, the first vibration characteristic data, and the first noise characteristic data, wherein the first vibration characteristic data refers to an abnormal vibration signal, such as: abnormal vibration frequency, abnormal amplitude, etc., and the first noise characteristic data refers to an abnormal noise signal, such as: abnormal noise frequency, abnormal noise tone, and abnormal noise intensity.

Data integration is performed on the multiple first initial warning indicator sets, wherein data integration refers to data classification and organization of the multiple first initial warning indicator sets, so as to facilitate subsequent warning indicator analysis and obtain multiple first warning indicator sets. Then, the first warning indicator with the highest frequency of occurrence in the multiple first warning indicator sets is extracted, for example: the temperature data, abnormal vibration signal, and abnormal noise signal with the highest frequency of occurrence, to obtain the first warning indicator set, and then the first warning indicator set is added to the multiple warning indicator sets to obtain multiple warning indicator sets, wherein the current data and the warning indicator sets correspond one to one, and by generating multiple warning indicator sets, support is provided for the next step of constructing a current-warning threshold database.

A current-warning threshold database is constructed based on the mapping relationship between current data and a set of warning indicators.

Specifically, multiple current intervals within the expected current threshold are set based on multiple current data, wherein the current interval is the current interval between two adjacent current data. Based on the decision tree principle, the current interval is used as the main node, and the warning indicator set corresponding to the current interval is used as the subsidiary node of the main node. Multiple current intervals and multiple warning indicator sets are used as filling data to form a current-warning threshold database.

By constructing a current-warning threshold database based on the principle of decision tree, it provides support for the next step of warning threshold matching, and at the same time can improve the accuracy and efficiency of warning threshold matching.

Further, step five of the present disclosure includes:

Based on industrial big data, multiple historical monitoring logs of the bus duct are retrieved, and multiple sample real-time temperatures and multiple sample temperature warning thresholds in the multiple historical monitoring logs are extracted, wherein the sample real-time temperature and the sample temperature warning threshold have a one-to-one correspondence;

Performing correlation analysis based on the real-time temperatures of the multiple samples and the temperature warning thresholds of the multiple samples to determine the temperature-threshold correlation coefficient;

Extracting the standard working temperature in the working environment data, and calculating the temperature deviation between the real-time temperature and the standard working temperature to obtain a temperature deviation value;

Performing a threshold deviation analysis on the temperature deviation value according to the temperature-threshold correlation coefficient to determine the threshold deviation value;

The temperature warning threshold is corrected by using the threshold deviation value to obtain the updated temperature warning threshold.

Specifically, the temperature warning threshold is corrected according to the real-time temperature to obtain an updated temperature warning threshold as follows: first, based on industrial big data, information retrieval is performed with the bus duct as the retrieval data to obtain multiple historical monitoring logs of the bus duct. Then, multiple sample real-time temperatures and multiple sample temperature warning thresholds in the multiple historical monitoring logs are extracted, wherein the sample real-time temperature and the sample temperature warning threshold have a one-to-one correspondence, wherein the sample real-time temperature refers to the ambient temperature when the bus duct is working.

An association analysis algorithm is used to perform an association analysis between the temperature and the warning threshold according to the real-time temperatures of the multiple samples and the temperature warning thresholds of the multiple samples. Commonly used association analysis algorithms include Apriori algorithm, FP-Growth algorithm, ECLAT algorithm, etc., which can be selected according to actual conditions to obtain a temperature-threshold correlation coefficient, where the temperature-threshold correlation coefficient refers to the impact of ambient temperature changes on the temperature threshold, for example: when the ambient temperature drops by 5 degrees Celsius, the temperature threshold drops by 1 degree Celsius, etc.

The standard operating temperature in the working environment data is extracted. The standard operating temperature is the temperature data set manually when constructing the bus duct twin model. Then, the temperature deviation between the real-time temperature and the standard operating temperature is calculated to obtain a temperature deviation value. Further, the temperature deviation value is subjected to a threshold deviation calculation based on the temperature-threshold correlation coefficient to generate a threshold deviation value. Finally, the temperature warning threshold is corrected based on the threshold deviation value, that is, the threshold deviation value is subtracted from the temperature warning threshold to obtain an updated temperature warning threshold.

By correcting the temperature warning threshold according to the real-time temperature, the accuracy of setting the temperature warning threshold can be further improved, thereby improving the accuracy of temperature warning judgment.

Further, step seven of the present disclosure includes:

Performing temperature deviation calculation on the updated temperature warning threshold and the temperature monitoring data to obtain a temperature deviation;

Performing a warning intensity analysis based on the temperature deviation to determine the warning intensity;

The warning intensity is input into a fault maintenance database for matching, an optimized maintenance plan is obtained, and the optimized maintenance plan is sent to corresponding fault maintenance personnel.

Specifically, before performing bus duct fault inspection and maintenance, first, the temperature deviation of the updated temperature warning threshold and the temperature monitoring data is calculated to obtain the temperature deviation, wherein the temperature deviation refers to the difference between the temperature monitoring data and the updated temperature warning threshold. Then, the warning intensity is analyzed according to the temperature deviation, wherein the greater the temperature deviation, the greater the warning intensity, and the warning intensity is obtained.

The warning intensity is input into the fault maintenance database for matching, an optimized maintenance plan is obtained, and the optimized maintenance plan is sent to the corresponding fault maintenance personnel, who perform bus duct fault maintenance and maintenance based on the optimized maintenance plan and the abnormal trough position coordinates.

The method for constructing the fault maintenance database is as follows: first, retrieve and obtain multiple bus duct maintenance logs, extract multiple historical warning intensities and corresponding multiple maintenance plans based on the multiple bus duct maintenance logs, and then cluster the multiple historical warning intensities to obtain multiple warning intensity intervals and corresponding multiple maintenance plan sets; comprehensively evaluate the multiple maintenance plans in turn through the intelligent expert system, and select the maintenance plan with the highest evaluation value as the optimized maintenance plan corresponding to the warning intensity interval; construct a fault maintenance database based on the mapping relationship between the warning intensity interval and the optimized maintenance plan.

By constructing a fault maintenance database to optimize the maintenance plan matching, the efficiency of obtaining the maintenance plan can be improved. At the same time, based on the optimized maintenance plan and the abnormal slot position coordinates, the corresponding slot is inspected and maintained, and the abnormal bus trough slot can be effectively handled in time to avoid major safety losses.

In summary, the bus duct fault diagnosis method for a photovoltaic energy storage system provided by the present disclosure has the following technical effects:

1. By dynamically adjusting the fault warning threshold based on photovoltaic power generation prediction, the accuracy of the fault warning threshold setting can be improved. At the same time, the real-time operation status of the bus duct can be warned by combining the fault warning thresholds of multiple dimensions, which can improve the accuracy and timeliness of the bus duct fault warning, so that the abnormal bus duct fault warning can be carried out in a timely and effective manner to avoid major safety losses.

2. By building a bus duct twin model based on digital twin technology, the authenticity and accuracy of the bus duct simulation test can be improved, thereby improving the accuracy and rationality of the early warning indicators.

3. By correcting the temperature warning threshold according to the real-time temperature, the accuracy of the temperature warning threshold setting can be further improved, thereby improving the accuracy of the temperature warning judgment.

Embodiment 2

Based on the same inventive concept as the bus duct fault diagnosis method for a photovoltaic energy storage system in the aforementioned embodiment, the present disclosure also provides a bus duct fault diagnosis system for a photovoltaic energy storage system, refer to FIG. 3 , the system comprises:

A current prediction module 11, which is used to transmit the light intensity and real-time temperature collected within a preset time window to a current prediction channel for current prediction and output a predicted current, wherein the current prediction channel is constructed based on a target photovoltaic device;

A current-early warning threshold database construction module 12, wherein the current-early warning threshold database construction module 12 is used to obtain an expected current threshold, perform multiple early warning tests on the bus duct under multiple current data within the expected current threshold according to multi-dimensional early warning indicators, and construct a current-early warning threshold database according to the test results;

A fault warning threshold matching module 13, wherein the fault warning threshold matching module 13 is used to input the predicted current into the current-warning threshold database, and match to obtain a fault warning threshold, wherein the fault warning threshold includes a temperature warning threshold, a vibration warning threshold, and a noise warning threshold;

A sensor monitoring data set receiving module 14, wherein the sensor monitoring data set receiving module 14 is used to receive a sensor monitoring data set of the bus duct, wherein the sensor monitoring data includes temperature monitoring data, vibration monitoring data and noise monitoring data, and each slot of the bus duct corresponds to a sensor monitoring data, wherein the sensor monitoring data is marked with a slot number;

A sensor monitoring data judgment module 15, the sensor monitoring data judgment module 15 is used to correct the temperature warning threshold according to the real-time temperature to obtain an updated temperature warning threshold, and judge the sensor monitoring data in the sensor monitoring data set based on the updated temperature warning threshold, the vibration warning threshold and the noise warning threshold;

The abnormal slot number set obtaining module 16 is used for extracting and abnormally marking the corresponding slot number when the temperature monitoring data does not meet the updated temperature warning threshold and/or the vibration monitoring data meets the vibration warning threshold and/or the noise monitoring data meets the noise warning threshold, so as to obtain an abnormal slot number set;

The bus duct fault inspection module 17 is used to locate and identify the abnormal duct body based on the abnormal duct body number set, send the abnormal duct body position coordinates to the nearest fault inspection personnel, and perform bus duct fault inspection and maintenance.

Furthermore, the current prediction module 11 in the system is also used for:

Obtaining basic indicator data of target photovoltaic equipment, wherein the basic indicator data includes equipment type, installed capacity, and photoelectric conversion efficiency;

Based on the industrial big data, the photovoltaic power generation related data is retrieved with the basic indicator data as the retrieval condition to obtain a plurality of sample power generation data, wherein the sample power generation data includes sample light intensity, sample temperature and sample power generation current;

performing data cleaning on the plurality of sample power generation data to obtain a plurality of standard sample power generation data;

The plurality of standard sample power generation data are used as training data, and supervised learning is performed on the current prediction channel constructed based on the BP neural network to obtain a current prediction channel that meets expected indicators.

Furthermore, the current prediction module 12 in the system is also used for:

The multi-dimensional early warning indicators include temperature indicators, vibration indicators and noise indicators;

Based on a preset analysis step length, the expected current threshold is divided into equal values to obtain a plurality of current data;

Constructing a bus duct twin model, and sequentially performing multiple early warning tests under multiple current data through the bus duct twin model to obtain multiple early warning indicator sets;

A current-warning threshold database is constructed based on the mapping relationship between current data and a set of warning indicators.

Furthermore, the current prediction module 12 in the system is also used for:

Obtaining equipment specification data, operation control parameters and working environment data of the bus duct;

In the visual simulation platform, simulation modeling of the bus duct equipment is performed based on the equipment specification data and the operation control parameters to obtain a twin model of the bus duct equipment;

The bus duct equipment twin model is environmentally configured based on the working environment data to obtain the bus duct twin model.

Furthermore, the current prediction module 12 in the system is also used for:

randomly selecting first current data from the plurality of current data;

Inputting the first current data into the bus duct twin model, performing multiple fault warning tests under a preset number threshold, and obtaining multiple first initial warning indicator sets, wherein the first initial warning indicator set includes a first temperature, a first vibration characteristic data, and a first noise characteristic data;

Performing data integration on the multiple first initial early warning indicator sets to obtain multiple first early warning indicator sets;

The first warning indicator with the highest occurrence frequency in the multiple first warning indicator sets is extracted to construct a first warning indicator set, and the first warning indicator set is added to the multiple warning indicator sets.

Furthermore, the current prediction module 15 in the system is also used for:

Based on industrial big data, multiple historical monitoring logs of the bus duct are retrieved, and multiple sample real-time temperatures and multiple sample temperature warning thresholds in the multiple historical monitoring logs are extracted, wherein the sample real-time temperature and the sample temperature warning threshold have a one-to-one correspondence;

Performing correlation analysis based on the real-time temperatures of the multiple samples and the temperature warning thresholds of the multiple samples to determine the temperature-threshold correlation coefficient;

Extracting the standard working temperature in the working environment data, and calculating the temperature deviation between the real-time temperature and the standard working temperature to obtain a temperature deviation value;

Performing a threshold deviation analysis on the temperature deviation value according to the temperature-threshold correlation coefficient to determine the threshold deviation value;

The temperature warning threshold is corrected by using the threshold deviation value to obtain the updated temperature warning threshold.

Furthermore, the current prediction module 17 in the system is also used for:

Performing temperature deviation calculation on the updated temperature warning threshold and the temperature monitoring data to obtain a temperature deviation;

Performing a warning intensity analysis based on the temperature deviation to determine the warning intensity;

The warning intensity is input into a fault maintenance database for matching, an optimized maintenance plan is obtained, and the optimized maintenance plan is sent to corresponding fault maintenance personnel.

Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The bus duct fault diagnosis method for photovoltaic energy storage system and the specific examples in the embodiment 1 of Figure 1 are also applicable to the bus duct fault diagnosis system for photovoltaic energy storage system in this embodiment. Through the detailed description of the bus duct fault diagnosis method for photovoltaic energy storage system, those skilled in the art can clearly know the bus duct fault diagnosis system for photovoltaic energy storage system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

The above description of the disclosed embodiments enables those skilled in the art to implement or use the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure belong to the scope of the present disclosure and its equivalent technology, the present disclosure is also intended to include these modifications and variations.

Claims (8)

1. A bus duct fault diagnosis method for a photovoltaic energy storage system, the method comprising:

Transmitting the illumination intensity and the real-time temperature acquired in a preset time window to a current prediction channel for current prediction, and outputting predicted current, wherein the current prediction channel is constructed based on target photovoltaic equipment;

Acquiring an expected current threshold, executing multiple times of early warning tests on bus ducts under a plurality of current data in the expected current threshold according to a multi-dimensional early warning index, and constructing a current-early warning threshold database according to test results;

Inputting the predicted current into the current-early warning threshold database, and matching to obtain a fault early warning threshold, wherein the fault early warning threshold comprises a temperature early warning threshold, a vibration early warning threshold and a noise early warning threshold;

receiving a sensing monitoring data set of the bus duct, wherein the sensing monitoring data comprises temperature monitoring data, vibration monitoring data and noise monitoring data, and each duct body of the bus duct corresponds to one sensing monitoring data, and the sensing monitoring data is provided with a duct body numbering mark;

Correcting the temperature early warning threshold according to the real-time temperature to obtain an updated temperature early warning threshold, and judging the sensing monitoring data in the sensing monitoring data set based on the updated temperature early warning threshold, the vibration early warning threshold and the noise early warning threshold respectively;

when the temperature monitoring data does not meet the updated temperature early warning threshold value and/or the vibration monitoring data meets the vibration early warning threshold value and/or the noise monitoring data meets the noise early warning threshold value, extracting and marking the corresponding groove body number abnormally to obtain an abnormal groove body number set;

and carrying out positioning identification based on the abnormal tank body number set, sending the position coordinates of the abnormal tank body to a fault maintainer closest to the abnormal tank body, and executing bus duct fault overhaul and maintenance.

2. The method of claim 1, wherein the current prediction channel is constructed based on a target photovoltaic device, comprising:

Basic index data of target photovoltaic equipment are obtained, wherein the basic index data comprise equipment types, installed capacity and photoelectric conversion efficiency;

Based on industrial big data, carrying out photovoltaic power generation related data retrieval by taking the basic index data as a retrieval condition to obtain a plurality of sample power generation data, wherein the sample power generation data comprises sample illumination intensity, sample temperature and sample power generation current;

performing data cleaning on the plurality of sample power generation data to obtain a plurality of standard sample power generation data;

Taking the plurality of standard sample power generation data as training data, and performing supervised learning on the current prediction channel constructed based on the BP neural network to obtain the current prediction channel meeting expected indexes.

3. The method of claim 1, wherein performing a plurality of pre-alarm tests on the bus duct under a plurality of current data within the expected current threshold according to the multi-dimensional pre-alarm indicator, and constructing a current-pre-alarm threshold database according to the test results, comprises:

The multi-dimensional early warning indexes comprise a temperature index, a vibration index and a noise index;

performing equivalence division on the expected current threshold value based on a preset analysis step length to obtain a plurality of current data;

Constructing a bus duct twin model, and sequentially executing multiple early warning tests under multiple current data through the bus duct twin model to obtain multiple early warning index sets;

And constructing a current-early warning threshold database based on the mapping relation between the current data and the early warning index set.

4. A method according to claim 3, wherein constructing a bus duct twinning model comprises:

acquiring equipment specification data, operation control parameters and working environment data of the bus duct;

in a visual simulation platform, simulation modeling of the bus duct equipment is carried out based on the equipment specification data and the operation control parameters, and a bus duct equipment twin model is obtained;

and carrying out environment configuration on the bus duct equipment twin model based on the working environment data to obtain the bus duct twin model.

5. The method of claim 3, wherein sequentially performing a plurality of early warning tests under a plurality of current data via the bus duct twinning model to obtain a plurality of sets of early warning indicators, comprises:

randomly selecting first current data from the plurality of current data;

Inputting the first current data into the bus duct twin model, and executing multiple fault early warning tests under a preset frequency threshold to obtain multiple first initial early warning index sets, wherein the first initial early warning index sets comprise first temperature, first vibration characteristic data and first noise characteristic data;

Data integration is carried out on the plurality of first initial early warning index sets to obtain a plurality of first early warning index sets;

Extracting first early warning indexes with highest occurrence frequency in the first early warning index sets, constructing a first early warning index set, and adding the first early warning index set into the first early warning index sets.

6. The method of claim 4, wherein correcting the temperature early warning threshold based on the real-time temperature to obtain an updated temperature early warning threshold comprises:

Based on industrial big data, searching a plurality of historical monitoring logs of the bus duct, and extracting a plurality of sample real-time temperatures and a plurality of sample temperature early warning thresholds in the historical monitoring logs, wherein the sample real-time temperatures and the sample temperature early warning thresholds have a one-to-one correspondence;

Performing correlation analysis based on the real-time temperatures of the plurality of samples and the temperature early warning thresholds of the plurality of samples, and determining a temperature-threshold correlation coefficient;

Extracting standard working temperature in the working environment data, and calculating the temperature deviation between the real-time temperature and the standard working temperature to obtain a temperature deviation value;

Performing threshold deviation analysis on the temperature deviation value according to the temperature-threshold correlation coefficient to determine a threshold deviation value;

and correcting the temperature early warning threshold value through the threshold value deviation value to obtain the updated temperature early warning threshold value.

7. The method of claim 1, further comprising, prior to performing bus duct troubleshooting and maintenance:

Performing temperature deviation calculation on the updated temperature early warning threshold and the temperature monitoring data to obtain temperature deviation;

performing early warning intensity analysis based on the temperature deviation to determine early warning intensity;

And inputting the early warning intensity into a fault overhaul database for matching to obtain an optimized overhaul scheme, and sending the optimized overhaul scheme to corresponding fault overhaul personnel.

8. A bus duct fault diagnosis system for a photovoltaic energy storage system, characterized by the steps for implementing the bus duct fault diagnosis method for a photovoltaic energy storage system according to any one of claims 1 to 7, the system comprising:

the current prediction module is used for transmitting the illumination intensity and the real-time temperature acquired in a preset time window to the current prediction channel for current prediction, outputting a predicted current, and constructing the current prediction channel based on target photovoltaic equipment;

the current-early warning threshold database construction module is used for acquiring an expected current threshold, executing multiple early warning tests on bus ducts under a plurality of current data in the expected current threshold according to multi-dimensional early warning indexes, and constructing a current-early warning threshold database according to test results;

the fault early warning threshold matching module is used for inputting the predicted current into the current-early warning threshold database, and matching to obtain a fault early warning threshold, wherein the fault early warning threshold comprises a temperature early warning threshold, a vibration early warning threshold and a noise early warning threshold;

The sensing monitoring data set receiving module is used for receiving a sensing monitoring data set of the bus duct, wherein the sensing monitoring data comprises temperature monitoring data, vibration monitoring data and noise monitoring data, each duct body of the bus duct corresponds to one sensing monitoring data, and the sensing monitoring data is provided with a duct body numbering mark;

The sensing monitoring data judging module is used for correcting the temperature early-warning threshold according to the real-time temperature to obtain an updated temperature early-warning threshold, and judging the sensing monitoring data in the sensing monitoring data set based on the updated temperature early-warning threshold, the vibration early-warning threshold and the noise early-warning threshold respectively;

The abnormal tank body number set obtaining module is used for extracting and marking the corresponding tank body number to obtain an abnormal tank body number set when the temperature monitoring data does not meet the updated temperature early warning threshold value and/or the vibration monitoring data meets the vibration early warning threshold value and/or the noise monitoring data meets the noise early warning threshold value;

And the bus duct fault maintenance module is used for carrying out positioning identification based on the abnormal tank body number set, sending the position coordinates of the abnormal tank body to a fault maintenance person closest to the abnormal tank body, and executing bus duct fault maintenance and maintenance.