Photovoltaic equipment state discrimination and fault diagnosis method based on AI

Abstract

The invention discloses a photovoltaic equipment state discrimination and fault diagnosis method based on AI, and relates to the technical field of fault diagnosis, and the method comprises the steps: step 1, photovoltaic equipment data acquisition, step 2, photovoltaic equipment state analysis, step 3, string fault inspection, and step 4, string fault processing. Whether each group string of the power plant has various fault abnormalities or not is judged, so that follow-up classified inspection is facilitated, inspection is performed by controlling the unmanned aerial vehicle, the open circuit range of the group string with the open circuit problem is determined, large-range troubleshooting required by a maintenance principal is avoided, the maintenance efficiency of the group string is improved, and the maintenance cost is reduced. And meanwhile, some solar cells with slight faults are checked, and the solar cells with the faults are timely positioned and maintained, so that the power generation efficiency of the string is improved, and the service life of the solar cells is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a photovoltaic equipment state identification and fault diagnosis method based on AI. Background Art

[0002] Since photovoltaic equipment uses solar energy to generate electricity and causes less damage to the environment, it is used as one of the main sources of power generation in more and more regions. In order to facilitate the management of solar panels in photovoltaic equipment, they are generally connected in series to obtain strings. However, once a solar panel has a problem, it is easy to cause harm to the entire string. Therefore, it is necessary to judge the status and diagnose the faults of the string.

[0003] Prior art, such as the invention patent application with announcement number: CN117951633B, discloses a photovoltaic power generation equipment fault diagnosis method and system, the method comprising: this invention collects and pre-processes the operating data of the photovoltaic power generation equipment to obtain a stable operating data set and an abnormal switching data set, performs feature extraction and cluster analysis on the merged data set, determines the association relationship between data items in the cluster, and based on the results of clustering and association analysis, uses feature engineering to extract new features of the data set, determines abnormal conditions during state switching of the photovoltaic power generation equipment, and timely traces and regulates the equipment that generates abnormal signals, thereby increasing the accuracy of identifying abnormal faults of the photovoltaic power generation equipment during the switching of operating modes, thereby improving the power generation efficiency.

[0004] Prior art, such as the invention patent application with announcement number: CN116777430A, discloses a line loss location and analysis method for a new power system, which includes: quickly locating the abnormal power loss position, greatly reducing costs and reducing waste of resources; at the same time, accurately locating the line fault point based on the conductor material, providing a strong reference basis for subsequent line maintenance and transformation; real-time monitoring of each load device of each terminal user, timely detection of faulty equipment and self-power-off processing, avoiding ineffective operation of the equipment, and avoiding the impact of faulty equipment on the entire power system, thereby improving the stability of the power system.

[0005] It can be seen from the above scheme that the current string status judgment and fault diagnosis methods lack certain attention to the positioning of the circuit breaking range of the string with the circuit breaking problem. During the use of the string, once a circuit breaking occurs, the entire string will not be able to generate current normally. The traditional circuit breaking analysis method generally only determines whether the string has a circuit breaking problem, but does not determine the circuit breaking range of the string, resulting in the need for the maintenance person in charge to conduct a large-scale investigation, thereby reducing the maintenance efficiency of the string. At the same time, there is a lack of attention to the use of drones to locate faulty solar cells. During the use of solar cells, some minor faults are prone to occur. If these faults are not maintained in time, the solar cells will further deteriorate, thereby reducing the power generation efficiency of the string and reducing the life of the solar cells. Summary of the invention

[0006] The purpose of the present invention is to provide a photovoltaic equipment state identification and fault diagnosis method based on AI, which solves the problems existing in the background technology.

[0007] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an AI-based photovoltaic equipment status judgment and fault diagnosis method, including: Step 1. Photovoltaic equipment data acquisition: Obtain the current value of each string belonging to the power plant at the current monitoring time point.

[0008] Step 2. PV equipment status analysis: Screen each disconnected string, each short-circuited string and each string to be analyzed belonging to the power plant, and power off each short-circuited string belonging to the power plant, and screen each abnormal string belonging to the power plant.

[0009] The specific screening method for screening abnormal strings belonging to the power plant is as follows: Step 2.1. String monitoring: obtaining the current value of each to-be-analyzed string belonging to the power plant at each monitoring time point within a detection time period.

[0010] Step 2.2. String abnormality analysis: Analyze the current dispersion coefficient of each string to be analyzed belonging to the power plant, obtain the current dispersion coefficient threshold from the local database, and if the current dispersion coefficient of a string to be analyzed belonging to the power plant is greater than the current dispersion coefficient threshold, mark the string to be analyzed as an abnormal string, thereby screening the abnormal strings belonging to the power plant.

[0011] Step 3. String fault inspection: The intelligent control drone inspects and analyzes the reserved circuit breaking detection points of each circuit breaking string belonging to the power plant and each solar cell of each abnormal string, thereby screening out each blocked solar cell, each hot spot solar cell and each faulty solar cell of each abnormal string belonging to the power plant, and obtaining the estimated line circuit breaking range of each circuit breaking string belonging to the power plant.

[0012] The specific method for conducting inspection and fault analysis is as follows: Step 3.1. UAV inspection: obtain the serial number of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant from the local database, and control the UAV to perform voltage tests on each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant in sequence, so as to obtain the voltage value of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant.

[0013] The current light intensity of the power plant is obtained, and the UAV is controlled to obtain the surface color image and infrared image of each solar cell of each abnormal string belonging to the power plant, so as to screen out each blocked solar cell, each hot spot solar cell and each analysis solar cell of each abnormal string belonging to the power plant, and the UAV is controlled to perform a low light source blocking test on each analysis solar cell of each abnormal string belonging to the power plant, so as to obtain the comparative current value, target current value and target temperature value of each monitoring point of each analysis solar cell of each abnormal string belonging to the power plant.

[0014] Step 3.2. Breaking string analysis: Analyze the estimated line breaking range of each breaking string belonging to the power plant.

[0015] Step 3.3. Abnormal string analysis: Evaluate the fault hazard coefficient of each analyzed solar cell of each abnormal string belonging to the power plant, so as to screen out each faulty solar cell of each abnormal string belonging to the power plant.

[0016] Step 4. String fault processing: each disconnected string belonging to the power plant and each abnormal string containing each hot spot solar cell is powered off, and the estimated line disconnection range of each blocked solar cell, each hot spot solar cell, each faulty solar cell and each disconnected string of each abnormal string belonging to the power plant is sent to the person in charge of the equipment.

[0017] The beneficial effects of the present invention are as follows: (1) Step 1 of the present invention is to obtain photovoltaic equipment data, by obtaining data of each string belonging to a power plant, thereby facilitating subsequent analysis.

[0018] (2) Step 2 of the present invention analyzes the status of photovoltaic equipment, by preliminarily monitoring the various strings belonging to the power plant, to determine whether various faults or abnormalities occur in the various strings belonging to the power plant, thereby facilitating subsequent classified inspections.

[0019] (3) Step 3 of the present invention is to inspect string faults by controlling drones to determine the range of the strings with circuit breakage problems, thereby avoiding the need for maintenance personnel to conduct large-scale inspections, thereby improving the maintenance efficiency of the strings. At the same time, the solar cells with some minor faults are inspected, and the faulty solar cells are located and maintained in time, thereby improving the power generation efficiency of the strings and ensuring the life of the solar cells.

[0020] (4) Step 4 of the present invention, string fault handling, allows the person in charge of the equipment to promptly understand the status and location of string problems, thereby facilitating subsequent maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Reference Figure 1 As shown, the present invention provides an AI-based photovoltaic equipment state identification and fault diagnosis method, including: Step 1. Photovoltaic equipment data acquisition: acquiring the current value of each string belonging to the power plant at the current monitoring time point.

[0025] In a specific embodiment, the current value of each string belonging to the power plant is obtained by a specific method: the current value of each string belonging to the power plant is obtained through a combiner box.

[0026] Step 1 of the present invention is to obtain photovoltaic equipment data, by obtaining the data of each string belonging to the power plant, so as to facilitate subsequent analysis.

[0027] Step 2. PV equipment status analysis: Screen each disconnected string, each short-circuited string and each string to be analyzed belonging to the power plant, and power off each short-circuited string belonging to the power plant, and screen each abnormal string belonging to the power plant.

[0028] In a specific embodiment of the present invention, the screening of each broken string, each short-circuited string and each to-be-analyzed string belonging to the power plant, the specific screening method is: obtaining the broken current value threshold and the short-circuit current value threshold from the local database, if the current value of a certain string belonging to the power plant is less than the broken current value threshold, then the string is marked as a broken string, otherwise, the current value of the string belonging to the power plant is compared with the short-circuit current value threshold, if the current value of the string belonging to the power plant is greater than the short-circuit current value threshold, then the string is marked as a short-circuited string, otherwise, the string is marked as a to-be-analyzed string, thereby screening each broken string, each short-circuited string and each to-be-analyzed string belonging to the power plant.

[0029] In a specific embodiment of the present invention, the specific screening method for screening abnormal strings belonging to the power plant is as follows: Step 2.1. String monitoring: obtaining the current value of each to-be-analyzed string belonging to the power plant at each monitoring time point within a detection time period.

[0030] In a specific embodiment, the current value of each to-be-analyzed group string belonging to the power plant at each monitoring time point is obtained by a specific method: the current value of each to-be-analyzed group string belonging to the power plant at each monitoring time point is obtained through a combiner box.

[0031] Step 2.2. String abnormality analysis: Analyze the current dispersion coefficient of each string to be analyzed belonging to the power plant, obtain the current dispersion coefficient threshold from the local database, and if the current dispersion coefficient of a string to be analyzed belonging to the power plant is greater than the current dispersion coefficient threshold, mark the string to be analyzed as an abnormal string, thereby screening the abnormal strings belonging to the power plant.

[0032] In a specific embodiment of the present invention, the current dispersion coefficient of each to-be-analyzed group string of the power plant is analyzed, and the specific analysis method is as follows: according to the current value a of each to-be-analyzed group string of the power plant at each monitoring time point xi , where x represents the number of each string to be analyzed, x = 1, 2, ..., y, y is a positive integer greater than 2, i represents the number of each monitoring time point, i = 1, 2, ..., j, j is a positive integer greater than 2, calculate the current dispersion coefficient of each string to be analyzed in the power plant Where y is the number of strings to be analyzed, and j is the number of monitoring time points.

[0033] Step 2 of the present invention is to analyze the status of photovoltaic equipment, by preliminarily monitoring the strings belonging to the power plant, to determine whether various faults or abnormalities occur in the strings belonging to the power plant, so as to facilitate subsequent classified inspections.

[0034] Step 3. String fault inspection: The intelligent control drone inspects and analyzes the reserved circuit breaking detection points of each circuit breaking string belonging to the power plant and each solar cell of each abnormal string, thereby screening out each blocked solar cell, each hot spot solar cell and each faulty solar cell of each abnormal string belonging to the power plant, and obtaining the estimated line circuit breaking range of each circuit breaking string belonging to the power plant.

[0035] In a specific embodiment of the present invention, the inspection and fault analysis are carried out, and the specific method is as follows: Step 3.1. UAV inspection: obtain the serial number of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant from the local database, and control the UAV to perform voltage tests on each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant in sequence, so as to obtain the voltage value of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant.

[0036] The current light intensity of the power plant is obtained, and the UAV is controlled to obtain the surface color image and infrared image of each solar cell of each abnormal string belonging to the power plant, so as to screen out each blocked solar cell, each hot spot solar cell and each analysis solar cell of each abnormal string belonging to the power plant, and the UAV is controlled to perform a low light source blocking test on each analysis solar cell of each abnormal string belonging to the power plant, so as to obtain the comparative current value, target current value and target temperature value of each monitoring point of each analysis solar cell of each abnormal string belonging to the power plant.

[0037] Step 3.2. Breaking string analysis: Analyze the estimated line breaking range of each breaking string belonging to the power plant.

[0038] Step 3.3. Abnormal string analysis: Evaluate the fault hazard coefficient of each analyzed solar cell of each abnormal string belonging to the power plant, so as to screen out each faulty solar cell of each abnormal string belonging to the power plant.

[0039] In a specific embodiment, the current light intensity of the power plant is obtained by a specific method: the current light intensity of the power plant is obtained through a meteorological management platform.

[0040] In a specific embodiment, the surface color image and infrared image of each solar cell of each abnormal string belonging to the power plant are obtained by a specific method: the surface color image and infrared image of each solar cell of each abnormal string belonging to the power plant are obtained by a camera of a drone.

[0041] In a specific embodiment, the screening of each blocked solar cell, each hot spot solar cell and each analysis solar cell of each abnormal string belonging to the power plant is performed by a specific screening method: based on the surface color image of each solar cell of each abnormal string belonging to the power plant, and by image recognition technology, it is judged whether there are any obstructions on the surface of each solar cell of each abnormal string belonging to the power plant; if so, the solar cell with the obstruction is marked as a blocked solar cell, thereby screening the blocked solar cells of each abnormal string belonging to the power plant; if not, based on the infrared image of each solar cell of each abnormal string belonging to the power plant and based on the existing hot spot analysis technology, it is judged whether there are hot spot problems on the surface of each solar cell of each abnormal string belonging to the power plant; if so, the solar cell with the hot spot problem is marked as a hot spot solar cell, thereby screening the hot spot solar cells of each abnormal string belonging to the power plant; if not, the remaining solar cells are marked as analysis solar cells, thereby screening the analysis solar cells of each abnormal string belonging to the power plant.

[0042] In a specific embodiment of the present invention, the low light source shading test is performed, and its specific testing method is: obtaining the shading duration and the preset test light intensity from the local database, controlling the sunshade of the drone to completely cover the solar cell within the shading duration, and after covering, turning on the test light in the sunshade to illuminate the solar cell with the preset test light intensity.

[0043] In a specific embodiment of the present invention, the estimated line circuit breaking range of each circuit breaker string belonging to the power plant is analyzed, and the specific analysis method is: obtain the circuit breaking voltage value threshold from the local database, if the voltage value of a reserved circuit breaking detection point of a circuit breaker string belonging to the power plant is less than the circuit breaking voltage value threshold, then the range between the reserved circuit breaking detection point and the previous reserved circuit breaking detection point of the circuit breaker string belonging to the power plant is used as the estimated line circuit breaking range, thereby analyzing the estimated line circuit breaking range of each circuit breaker string belonging to the power plant.

[0044] In a specific embodiment of the present invention, the fault hazard coefficient of each analyzed solar cell of each abnormal group string belonging to the power plant is evaluated, and the specific evaluation method is: according to the comparison current value and target current value of each analyzed solar cell of each abnormal group string belonging to the power plant, the current abnormal coefficient δ of each analyzed solar cell of each abnormal group string belonging to the power plant is analyzed. np , wherein n represents the number of each abnormal string, n=1,2,...,m, m is a positive integer greater than 2, and p represents the number of each analyzed solar cell, p=1,2,...,q, q is a positive integer greater than 2.

[0045] According to the target temperature value of each monitoring point of each analyzed solar cell of each abnormal group string belonging to the power plant at each monitoring time point, the temperature anomaly coefficient ε of each analyzed solar cell of each abnormal group string belonging to the power plant is analyzed. np .

[0046] Calculate the failure hazard coefficient φ of each analyzed solar cell of each abnormal string in the power plant np =δ np +ε np .

[0047] In a specific embodiment of the present invention, the current abnormality coefficient of each analyzed solar cell of each abnormal group string belonging to the power plant is analyzed, and the specific analysis method is: obtain the relationship function between the light intensity and the output current value of the solar cell from the local database, and calculate the estimated output current value difference b of each analyzed solar cell of each abnormal group string belonging to the power plant when the light intensity of the power plant changes from the current light intensity of the power plant to the preset test light intensity according to the current light intensity of the power plant and the preset test light intensity. np .

[0048] According to the comparative current value c of each analyzed solar cell of each abnormal string in the power plant np , target current value d np , calculate the current abnormal coefficient of each analyzed solar cell of each abnormal string belonging to the power plant

[0049] It should be noted that the comparison current value is the total current value of each abnormal string belonging to the power plant before the low light source shading test is carried out on each analyzed solar cell, the target current value is the total current value of each abnormal string belonging to the power plant after the low light source shading test is carried out on each analyzed solar cell, and the target temperature value is the temperature value of each monitoring point of each abnormal string belonging to the power plant after the low light source shading test is carried out on each analyzed solar cell.

[0050] In a specific embodiment of the present invention, the temperature anomaly coefficient of each analyzed solar cell of each abnormal group string belonging to the power plant is analyzed, and the specific analysis method is: according to the infrared image of each solar cell of each abnormal group string belonging to the power plant, the initial temperature value f of each solar cell of each abnormal group string belonging to the power plant at each monitoring point is extracted. npt , where t represents the number of each monitoring point, t=1,2,...,s, and s is a positive integer greater than 2.

[0051] The use time of each solar cell of each abnormal group string of the power plant is obtained from the local database, and the ideal temperature drop value corresponding to each initial temperature value after the solar cell in each use time interval is tested for low light source shielding, and the ideal temperature drop value g of each solar cell of each abnormal group string of the power plant is mapped. np .

[0052] According to the target temperature value k of each monitoring point of each analyzed solar cell of each abnormal string belonging to the power plant npt , calculate the temperature anomaly coefficient of each analyzed solar cell of each abnormal string belonging to the power plant Where e represents a natural constant.

[0053] Step 3 of the present invention is to inspect string faults by controlling a drone to determine the circuit breaking range of the string with the circuit breaking problem, thereby avoiding the need for a large-scale investigation by the person in charge of maintenance, thereby improving the maintenance efficiency of the string. At the same time, the solar cells with some minor faults are investigated, and the faulty solar cells are located and maintained in time, thereby improving the power generation efficiency of the string and ensuring the life of the solar cells.

[0054] Step 4. String fault processing: each disconnected string belonging to the power plant and each abnormal string containing each hot spot solar cell is powered off, and the estimated line disconnection range of each blocked solar cell, each hot spot solar cell, each faulty solar cell and each disconnected string of each abnormal string belonging to the power plant is sent to the person in charge of the equipment.

[0055] Step 4 of the present invention is to handle string failures, so that the person in charge of the equipment can timely grasp the status and location of string problems, thereby facilitating subsequent maintenance.

[0056] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A photovoltaic equipment state identification and fault diagnosis method based on AI, characterized in that: include: Step 1. Photovoltaic equipment data acquisition: obtain the current value of each string belonging to the power plant at the current monitoring time point; Step 2. PV equipment status analysis: Screen each disconnected string, each short-circuited string and each string to be analyzed belonging to the power plant, and power off each short-circuited string belonging to the power plant, and screen each abnormal string belonging to the power plant; Step 3. String fault inspection: The intelligent control drone inspects and analyzes the reserved circuit breaking detection points of each circuit breaking string belonging to the power plant and each solar cell of each abnormal string, thereby screening out each blocked solar cell, each hot spot solar cell and each faulty solar cell of each abnormal string belonging to the power plant, and obtaining the estimated line circuit breaking range of each circuit breaking string belonging to the power plant; Step 4. String fault processing: each disconnected string belonging to the power plant and each abnormal string containing each hot spot solar cell is powered off, and the estimated line disconnection range of each blocked solar cell, each hot spot solar cell, each faulty solar cell and each disconnected string of each abnormal string belonging to the power plant is sent to the person in charge of the equipment.

2. According to claim 1, a photovoltaic equipment state identification and fault diagnosis method based on AI is characterized in that: The specific screening method of screening each disconnected string, each short-circuited string and each to-be-analyzed string belonging to the power plant is as follows: The breaking current value threshold and the short-circuit current value threshold are obtained from the local database. If the current value of a certain string belonging to the power plant is less than the breaking current value threshold, the string is marked as a breaking string. Otherwise, the current value of the string belonging to the power plant is compared with the short-circuit current value threshold. If the current value of the string belonging to the power plant is greater than the short-circuit current value threshold, the string is marked as a short-circuit string. Otherwise, the string is marked as a string to be analyzed, thereby screening the breaking strings, short-circuit strings and strings to be analyzed belonging to the power plant.

3. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 1 is characterized in that: The specific screening method for screening abnormal strings belonging to the power plant is as follows: Step 2.

1. String monitoring: obtaining the current value of each string to be analyzed belonging to the power plant at each monitoring time point within the detection time period; Step 2.

2. String abnormality analysis: Analyze the current dispersion coefficient of each string to be analyzed belonging to the power plant, obtain the current dispersion coefficient threshold from the local database, and if the current dispersion coefficient of a string to be analyzed belonging to the power plant is greater than the current dispersion coefficient threshold, mark the string to be analyzed as an abnormal string, thereby screening the abnormal strings belonging to the power plant.

4. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 3 is characterized in that: The specific analysis method of analyzing the current dispersion coefficient of each string to be analyzed in the power plant is as follows: According to the current value a of each string to be analyzed in the power plant at each monitoring time point xi , where x represents the number of each string to be analyzed, x = 1, 2, ..., y, y is a positive integer greater than 2, i represents the number of each monitoring time point, i = 1, 2, ..., j, j is a positive integer greater than 2, calculate the current dispersion coefficient of each string to be analyzed in the power plant Where y is the number of strings to be analyzed, and j is the number of monitoring time points.

5. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 1, characterized in that: The specific method of performing inspection and fault analysis is as follows: Step 3.

1. UAV inspection: obtain the serial number of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant from the local database, control the UAV to perform voltage tests on each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant in sequence, so as to obtain the voltage value of each reserved circuit breaker detection point of each circuit breaker string belonging to the power plant; The current light intensity of the power plant is obtained, and the UAV is controlled to obtain the surface color image and infrared image of each solar cell of each abnormal string belonging to the power plant, so as to screen out each blocked solar cell, each hot spot solar cell and each analysis solar cell of each abnormal string belonging to the power plant, and the UAV is controlled to perform a low light source blocking test on each analysis solar cell of each abnormal string belonging to the power plant, so as to obtain the comparison current value, target current value and target temperature value of each monitoring point of each analysis solar cell of each abnormal string belonging to the power plant; Step 3.

2. Analysis of disconnected strings: Analyze the estimated disconnection range of each disconnected string belonging to the power plant; Step 3.

3. Abnormal string analysis: Evaluate the fault hazard coefficient of each analyzed solar cell of each abnormal string belonging to the power plant, so as to screen out each faulty solar cell of each abnormal string belonging to the power plant.

6. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 5, characterized in that: The specific test method for the low light source occlusion test is as follows: The shielding duration and the preset test light intensity are obtained from the local database, and the sunshade of the drone is controlled to completely shield the solar cell within the shielding duration. After shielding, the test light in the sunshade is turned on to irradiate the solar cell with the preset test light intensity.

7. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 5, characterized in that: The specific analysis method for analyzing the estimated line breaking range of each breaking string belonging to the power plant is as follows: The circuit breaker voltage value threshold is obtained from the local database. If the voltage value of a reserved circuit breaker detection point of a circuit breaker string belonging to a power plant is less than the circuit breaker voltage value threshold, the range between the reserved circuit breaker detection point and the previous reserved circuit breaker detection point of the circuit breaker string belonging to the power plant is taken as the estimated line circuit breaker range, thereby analyzing the estimated line circuit breaker range of each circuit breaker string belonging to the power plant.

8. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 6, characterized in that: The specific evaluation method for evaluating the failure hazard coefficient of each analyzed solar cell of each abnormal string belonging to the power plant is as follows: According to the comparison current value and target current value of each analyzed solar cell of each abnormal group string of the power plant, the current abnormal coefficient δ of each analyzed solar cell of each abnormal group string of the power plant is analyzed. np , where n represents the number of each abnormal string, n=1,2,...,m, m is a positive integer greater than 2, p represents the number of each analyzed solar cell, p=1,2,...,q, q is a positive integer greater than 2; According to the target temperature value of each monitoring point of each analyzed solar cell of each abnormal group string belonging to the power plant at each monitoring time point, the temperature anomaly coefficient ε of each analyzed solar cell of each abnormal group string belonging to the power plant is analyzed. np ; Calculate the failure hazard coefficient φ of each analyzed solar cell of each abnormal string in the power plant np =δ np +ε np .

9. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 8, characterized in that: The specific analysis method of the abnormal current coefficient of each analyzed solar cell of each abnormal string belonging to the analyzed power plant is as follows: The relationship function between the light intensity and output current value of the solar cell is obtained from the local database, and based on the current light intensity of the power plant and the preset test light intensity, the estimated output current value difference b of each analyzed solar cell of each abnormal group string belonging to the power plant is calculated when the current light intensity of the power plant changes to the preset test light intensity. np ; According to the comparative current value c of each analyzed solar cell of each abnormal string in the power plant np , target current value d np , calculate the current abnormal coefficient of each analyzed solar cell of each abnormal string belonging to the power plant 10. The method for photovoltaic equipment state identification and fault diagnosis based on AI according to claim 8, characterized in that: The specific analysis method of the temperature anomaly coefficient of each analyzed solar cell of each abnormal string belonging to the analyzed power plant is as follows: According to the infrared images of each solar cell of each abnormal string belonging to the power plant, the initial temperature value f of each solar cell of each abnormal string belonging to the power plant at each monitoring point is extracted. npt , where t represents the number of each monitoring point, t = 1, 2, ..., s, s is a positive integer greater than 2; The use time of each solar cell of each abnormal group string of the power plant is obtained from the local database, and the ideal temperature drop value corresponding to each initial temperature value after the solar cell in each use time interval is tested for low light source shielding, and the ideal temperature drop value g of each solar cell of each abnormal group string of the power plant is mapped. np ; According to the target temperature value k of each monitoring point of each analyzed solar cell of each abnormal string belonging to the power plant npt , calculate the temperature anomaly coefficient of each analyzed solar cell of each abnormal string belonging to the power plant Where e represents a natural constant.

Images