Household photovoltaic fault state estimation method based on multi-dimensional data fusion

Abstract

According to the multi-dimensional data fusion-based household photovoltaic fault state estimation method, the construction operation of a household photovoltaic three-dimensional distribution electronic map can be carried out through a three-dimensional modeling assembly, and the household photovoltaic can be divided into different areas through the position information of the household photovoltaic, so that the fault state of the household photovoltaic can be estimated. The consistency of the meteorological condition and the output of the household photovoltaic in the sub-region is realized; according to the method, the photovoltaic current value of the household photovoltaic module can be calculated according to the set power of the household photovoltaic module and the actual generating capacity data of the household photovoltaic module, so that the fault household photovoltaic module can be preliminarily judged, and then the fault household photovoltaic module can be judged based on the marked historical generating capacity data of the household photovoltaic module through a normal distribution algorithm and a gradient descent method. According to the method, the preset power generation normal coefficient is determined, then the preset power generation interval of the household photovoltaic module is calculated according to the preset power generation normal coefficient, and finally the preset power generation interval is compared with the actual photo-generated current value, so that the final judgment of the fault household photovoltaic module is realized, and the method has the advantages of being accurate, efficient, convenient and reasonable.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of new energy power generation operation and management, and particularly relates to a household photovoltaic fault state estimation method based on multi-dimensional data fusion. BACKGROUND

[0002] In recent years, with the rapid growth of the photovoltaic industry, clean energy has attracted people's attention; household photovoltaic is a kind of distributed photovoltaic, that is, photovoltaic cell panels are placed on the top floor of a family residence or in a courtyard, and small power or micro-inverters are used for commutation, and then the new energy is directly used for power use, and the excess power can also be integrated into the power grid; however, with the increase in the number of household photovoltaics, various faults that occur when household photovoltaics are used also come; therefore, in order to facilitate the management of household photovoltaics, timely detection of faults and repair, it is necessary to develop a precise, efficient and convenient and reasonable household photovoltaic fault state estimation method based on multi-dimensional data fusion. SUMMARY

[0003] The purpose of the present application is to overcome the shortcomings of the prior art and provide a precise, efficient and convenient and reasonable household photovoltaic fault state estimation method based on multi-dimensional data fusion.

[0004] The purpose of the present application is achieved by a household photovoltaic fault state estimation method based on multi-dimensional data fusion, comprising the following steps: Step 1: according to the operation data and distribution map of the household photovoltaic components, a three-dimensional modeling component is used to construct a household photovoltaic three-dimensional distribution electronic map; Step 2: based on the geographic latitude and longitude coordinates, a plurality of regional boundary lines are drawn on the household photovoltaic three-dimensional distribution electronic map according to the position information, thereby realizing the partitioning of the household photovoltaics; Step 3: the position information, actual power generation and model of the household photovoltaic components in each sub-region after partitioning are collected and obtained, and based on the sub-region, the data are arranged and displayed in the form of a table, thereby forming a data calculation table of the sub-region, then based on the model of the photovoltaic components, the set power data of the photovoltaic components are filled in the data calculation table, and finally based on the set power of the household photovoltaic components and the actual power generation data, the photogenerated current value of the household photovoltaic components is calculated; Step 4: the photogenerated current values of the household photovoltaic components in the same region calculated in step 3 are compared and analyzed, and the household photovoltaic components with excessively large or small photogenerated current values are marked in yellow on the household photovoltaic three-dimensional distribution electronic map; Step 5: The historical power generation data of the yellow marked household photovoltaic module is retrieved in the historical database, then the maximum likelihood estimation analysis is carried out according to the historical power generation data by normal distribution algorithm and gradient descent method, the normal upper limit coefficient, the normal lower limit coefficient and the preset normal coefficient of power generation are determined, finally the preset power generation interval of the household photovoltaic module is calculated according to the preset normal coefficient of power generation, finally the preset power generation interval is compared with the actual photoelectric current value, if the difference is too large, it is judged as fault, the warning module sends warning signal at the same time, and the household photovoltaic module is marked with red in the household photovoltaic three-dimensional distribution electronic map; Step 6: Based on the household photovoltaic three-dimensional distribution electronic map, the position information of the red marked household photovoltaic is obtained, and the position information is integrated with the model information of the household photovoltaic, so as to form the maintenance work order, then the maintenance work order is distributed at the same time, and the nearby operation and maintenance personnel are arranged to carry out on-site fault diagnosis and maintenance.

[0005] Further, the operation data of the household photovoltaic module in step 1 includes: the installation address, the installation latitude and longitude, the MPPT voltage, the line voltage, the phase voltage and the power station power of the photovoltaic module.

[0006] Further, in step 2, the household photovoltaic is divided into several regions, and in the divided sub-regions, the meteorological conditions and the output of the household photovoltaic have consistency.

[0007] Further, in step 5, the warning module can send the warning signal according to the preset interval time, if the confirmation feedback instruction from the operation end is received, the sending is stopped, otherwise the sending will continue.

[0008] The beneficial effects of the present application are as follows: through the three-dimensional modeling component and based on the operation data and distribution map of the household photovoltaic component, the household photovoltaic three-dimensional distribution electronic map can be constructed, in this way, the distribution of the household photovoltaic component can be observed, and the convenience of subsequent fault household photovoltaic component marking operation is increased; through the position information of the household photovoltaic, the household photovoltaic can be partitioned, and the meteorological conditions and output of the household photovoltaic in the sub-region are consistent; through the set power of the household photovoltaic component and the actual power generation data, the photogenerated current value of the household photovoltaic component is calculated, and the photogenerated current values of the household photovoltaic components in the same region calculated are compared and analyzed, the household photovoltaic components with too large or too small photogenerated current values are marked with yellow in the household photovoltaic three-dimensional distribution electronic map, the preliminary determination of the fault household photovoltaic is realized, then, based on the historical power generation data of the marked household photovoltaic, and through the normal distribution algorithm and the gradient descent method, the historical power generation data is analyzed by maximum likelihood estimation, the normal upper limit coefficient of power generation, the normal lower limit coefficient of power generation and the preset normal coefficient of power generation are determined, the preset power generation interval of the household photovoltaic component is calculated according to the preset normal coefficient of power generation, and finally the preset power generation interval is compared with the actual photogenerated current value, so that the final determination of the fault household photovoltaic is realized, in this way, the fault household photovoltaic can be positioned efficiently and accurately, the labor intensity of the operation and maintenance personnel is greatly reduced, and through the secondary fault determination, the accuracy of use is increased; in general, the present application has the advantages of precision, efficiency, convenience and reasonableness. DETAILED DESCRIPTION

[0009] The present application will be further described below.

[0010] Embodiment: A household photovoltaic fault state estimation method based on multi-dimensional data fusion, comprising the following steps: Step 1: according to the operation data and distribution map of the household photovoltaic component, a three-dimensional modeling component is used to construct a household photovoltaic three-dimensional distribution electronic map; wherein the operation data of the household photovoltaic component includes: the installation address, installation longitude and latitude, MPPT voltage, line voltage, phase voltage and power station power of the photovoltaic component; Step 2: based on the geographic longitude and latitude coordinates, a plurality of regional boundary lines are drawn on the household photovoltaic three-dimensional distribution electronic map according to the position information, so as to realize the partitioning of the household photovoltaic, wherein in the partitioned sub-region, the meteorological conditions and the output of the household photovoltaic are consistent; Step 3: The position information, actual power generation and model of the household photovoltaic module in each sub-region are collected and obtained, and the data are arranged and displayed in the form of a table based on the sub-region, thereby forming a data calculation table of the sub-region. Then, the set power data of the photovoltaic module are filled in the data calculation table based on the model of the photovoltaic module. Finally, the photogenerated current value of the household photovoltaic module is calculated based on the set power of the household photovoltaic module and the actual power generation data. Step 4: The photogenerated current value of the household photovoltaic module in the same region calculated in step 3 is compared and analyzed, and the household photovoltaic module with too large or too small photogenerated current value is marked with yellow in the household photovoltaic three-dimensional distribution electronic map. Step 5: The historical power generation data of the household photovoltaic module marked with yellow are retrieved in the historical database. Then, the maximum likelihood estimation analysis is performed according to the historical power generation data by using the normal distribution algorithm and the gradient descent method to determine the upper limit coefficient, the lower limit coefficient and the preset normal coefficient of power generation. Finally, the preset power generation interval of the household photovoltaic module is calculated according to the preset normal coefficient of power generation. Finally, the preset power generation interval is compared with the actual photogenerated current value. If the difference is too large, it is determined as a fault, and the warning module sends a warning signal while marking the household photovoltaic module with red in the household photovoltaic three-dimensional distribution electronic map. The warning module can send a warning signal according to the preset interval time. If a confirmation feedback instruction is received from the operation end, the sending is stopped. Otherwise, the sending will continue. Step 6: Based on the household photovoltaic three-dimensional distribution electronic map, the position information of the household photovoltaic module marked with red is obtained, and the position information is integrated with the model information of the household photovoltaic module, thereby forming a maintenance work order. Then, the maintenance work order is distributed, and the nearby operation and maintenance personnel are arranged to conduct on-site fault diagnosis and maintenance.

[0011] The application is used, first, through the three-dimensional modeling component, and based on the operation data and distribution map of the household photovoltaic component, the household photovoltaic three-dimensional distribution electronic map construction operation is carried out, wherein the operation data of the household photovoltaic component includes: the installation address, the installation longitude and latitude of the photovoltaic component, the MPPT voltage, the line voltage, the phase voltage and the power station power; then, based on the position information, a plurality of regional boundary lines are drawn on the household photovoltaic three-dimensional distribution electronic map by using the geographic longitude and latitude coordinates, so as to realize the partition of the household photovoltaic, wherein the meteorological condition and the output of the household photovoltaic in the divided sub-area are consistent; then, the position information, the actual power generation and the model of the household photovoltaic component in each sub-area after division are collected and acquired, and the data calculation table of the sub-area is established based on the sub-area as the reference, then, based on the model of the photovoltaic component, the set power data of the photovoltaic component in the data calculation table is filled, and then, based on the set power of the household photovoltaic component and the actual power generation data, the photo-generated current value of the household photovoltaic component is calculated, and after the calculation is completed, the photo-generated current value of the household photovoltaic component in the same area is compared and analyzed, wherein the household photovoltaic component with too large or too small photo-generated current value is marked with yellow in the household photovoltaic three-dimensional distribution electronic map; after the above operation is completed, the historical power generation data of the household photovoltaic component marked with yellow in the historical database is retrieved, and the maximum likelihood estimation analysis is carried out according to the historical power generation data by using the normal distribution algorithm and the gradient descent method to determine the upper limit coefficient of normal power generation, the lower limit coefficient of normal power generation and the preset normal power generation coefficient, then, the preset power generation interval of the household photovoltaic component is calculated according to the preset normal power generation coefficient, and the preset power generation interval is compared with the actual photo-generated current value, if the difference is too large, it is determined as a fault, the warning module sends a warning signal at the same time, and the household photovoltaic is marked with red in the household photovoltaic three-dimensional distribution electronic map, in this process, the warning module can send the warning signal according to the preset interval time, if the confirmation feedback instruction from the operation end is received, the sending is stopped, otherwise the sending will continue; finally, based on the household photovoltaic three-dimensional distribution electronic map, the position information of the household photovoltaic marked with red is obtained, and the position information is integrated with the model information of the household photovoltaic, so as to form a maintenance work order, then, the maintenance work order is distributed at the same time, and the nearby operation and maintenance personnel are arranged to carry out on-site fault elimination and maintenance; by using this way, the application can efficiently and accurately locate the fault household photovoltaic, greatly reduces the labor intensity of the operation and maintenance personnel, and through the secondary fault determination method, the accuracy of use is increased; in general, the application has the advantages of high accuracy, high efficiency, convenience and reasonableness.

[0012] The above examples are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application is described in detail with reference to the above examples, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the scope of the claims of the present application.

Claims

1. A method for estimating the fault state of residential photovoltaic systems using multi-dimensional data fusion, characterized in that, Includes the following steps: Step 1: Based on the operating data and distribution map of the residential photovoltaic modules, construct a 3D electronic map of the residential photovoltaic distribution using 3D modeling components; Step 2: Based on geographical latitude and longitude coordinates, draw multiple regional boundary lines on the three-dimensional electronic map of household photovoltaic distribution according to location information, thereby realizing the zoning of household photovoltaic areas; Step 3: Collect and obtain data such as the location information, actual power generation, and model of the household photovoltaic modules in each sub-region after division. Based on the sub-region, organize and display the data in the form of a table to form a data calculation table for the sub-region. Then, based on the model of the photovoltaic module, fill in the set power data of the photovoltaic module in the data calculation table. Finally, based on the set power and actual power generation data of the household photovoltaic module, calculate the photocurrent value of the household photovoltaic module. Step 4: Compare and analyze the photovoltaic current values ​​of household photovoltaic modules in the same area calculated in Step 3, and mark household photovoltaic modules with excessively large or small photovoltaic current values ​​in yellow on the three-dimensional distribution electronic map of household photovoltaics. Step 5: Retrieve the historical power generation data of the yellow-marked residential photovoltaic modules from the historical database. Then, using the normal distribution algorithm and gradient descent method, perform maximum likelihood estimation analysis based on the historical power generation data to determine the upper limit coefficient, lower limit coefficient, and preset normal power generation coefficient. Finally, calculate the preset power generation range of the residential photovoltaic module based on the preset normal power generation coefficient. Finally, compare the preset power generation range with the actual photovoltaic current value. If the difference is too large, it is judged as a fault. At the same time, the alarm module issues an alarm signal and marks the residential photovoltaic module in red on the three-dimensional distribution electronic map of residential photovoltaics. Step 6: Based on the 3D electronic map of residential photovoltaic distribution, obtain the location information of the residential photovoltaic units marked in red, and integrate the location information with the model information of the residential photovoltaic units to form a maintenance work order. Then, while dispatching the maintenance work order, promptly arrange nearby operation and maintenance personnel to conduct on-site fault diagnosis and repair.

2. The method for estimating the fault state of residential photovoltaic systems using multi-dimensional data fusion as described in claim 1, characterized in that: The operating data of the residential photovoltaic modules in step 1 includes: the installation address of the photovoltaic modules, the installation latitude and longitude, MPPT voltage, line voltage, phase voltage and power station power.

3. The method for estimating the fault state of residential photovoltaic systems using multi-dimensional data fusion as described in claim 1, characterized in that: In step 2, the household photovoltaic system is divided into zones, and within each zone, the meteorological conditions and the output of the household photovoltaic system are consistent.

4. The method for estimating the fault state of residential photovoltaic systems using multi-dimensional data fusion as described in claim 1, characterized in that: In step 5, the warning module can send warning signals at preset intervals. If it receives a confirmation feedback instruction from the operating terminal, it will stop sending; otherwise, it will continue sending.