Optical storage equipment fault diagnosis system and method based on cloud platform

Abstract

The invention discloses an optical storage equipment fault diagnosis system and method based on a cloud platform. The system comprises photovoltaic equipment, energy storage equipment, a cleaning robot, a field monitoring machine, an equipment management server, a cloud platform server and an operation and maintenance terminal. The cleaning robot comprises a cleaning module, a driving module, a visual detection device and an environment sensor; the visual detection device is used for acquiring an environment image of the photovoltaic equipment and judging whether shadow shielding exists or not; the environment sensor is used for acquiring environment parameters of the energy storage equipment; the equipment management server comprises a photovoltaic data module, an energy storage data module and a robot data module; and the cloud platform server comprises a region division module, a correction compensation module, a route planning module and a model diagnosis module. According to the invention, a plurality of optical storage subsystems in one area are comprehensively diagnosed, the collected electrical parameters are corrected and compensated, data are comprehensively collected through the cleaning robot, and the diagnosis efficiency and the diagnosis accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and particularly to a fault diagnosis system and method for a photovoltaic energy storage device based on a cloud platform. Background Art

[0002] The increasing energy consumption, especially the extensive use of fossil fuels such as coal and oil, has brought severe challenges to the environment and the global climate. As a rapidly developing green and renewable energy source in recent years, the photovoltaic energy storage system has many advantages such as pollution-free, multiple carriers, and convenient utilization, and can be widely applied to various places such as factories, agricultural greenhouses, large buildings, and new energy vehicles. However, since the photovoltaic energy storage system contains many devices, the correlation between devices is strong, and the working scenarios are mostly outdoors, and its working environment is complex and harsh, so it is inevitable to have faults. How to diagnose the faults of the photovoltaic energy storage system is a technical problem that needs to be solved urgently.

[0003] In practical applications, the fault diagnosis of the photovoltaic energy storage system is usually manual diagnosis, which usually requires professionals to go to the site for information collection and analysis and judgment, consuming a lot of manpower and material resources, with low diagnosis efficiency, and it is difficult to detect the faults of the photovoltaic energy storage system in time. In the prior art, there are also literatures that diagnose the faults of the photovoltaic energy storage system through a machine learning model. However, the correlation and coupling between the devices of the photovoltaic energy storage system are strong, and the working environment is complex and changeable. If a large amount of data collected from the photovoltaic energy storage system is directly used as the input of the machine learning model, it will lead to problems such as low diagnosis efficiency, many interference factors, and inaccurate diagnosis. Moreover, due to the good development momentum of the photovoltaic energy storage system, a large number of photovoltaic energy storage systems will appear in all walks of life in the foreseeable future. Most of the existing fault diagnosis methods for the photovoltaic energy storage system are for individual photovoltaic energy storage systems, and there are few methods for comprehensive diagnosis and analysis of multiple photovoltaic energy storage subsystems in a region.

[0004] For example, document CN116455324A discloses an online fault diagnosis method and system for a photovoltaic energy storage system. The method and system include: collecting the original data of the photovoltaic energy storage system; using data-driven and convolutional neural network (CNN) to extract features from the original data to obtain feature data; and performing online fault diagnosis on the photovoltaic energy storage system according to the feature data. This document can improve the accuracy of online fault diagnosis for the photovoltaic energy storage system. However, the key steps of this document lie in what kind of original data to collect and what specific method to use to extract what kind of feature data. Such specific choices will seriously affect the efficiency and accuracy of fault diagnosis. This invention has the problem of difficult selection and processing of model input data. Another example is document CN218335427U, which discloses an off-grid photovoltaic energy storage power system health diagnosis device, including: a wall-mounted housing with a door panel; a control module installed in a rack-mounted housing for data processing and communication; an HMI unit installed on the door panel and connected to the control module for providing users to directly query and set data on the terminal; the control module includes a monitoring unit, a collection unit, and a storage unit. The monitoring unit is connected to the collection unit, and the storage unit is connected to the monitoring unit and the collection unit. The sensor unit is used to collect data information and upload it to the collection unit for processing. The communication unit is used to establish communication between the monitoring unit and the background host computer. This document has the characteristics of small volume, light weight, easy placement, and simple operation, and can achieve the effect of online continuous automatic monitoring and early warning. However, the object of this document is a single off-grid photovoltaic energy storage system, and it does not perform comprehensive diagnostic analysis on multiple photovoltaic energy storage subsystems in a region. Summary of the Invention

[0005] Object of the Invention: Aiming at the above problems, the present invention proposes a cloud platform-based fault diagnosis system and method for photovoltaic energy storage devices.

[0006] Technical Solution:

[0007] In a first aspect, the present invention proposes a cloud platform-based fault diagnosis system for photovoltaic energy storage devices, including: photovoltaic devices, energy storage devices, cleaning robots, on-site monitoring machines, equipment management servers, cloud platform servers, and operation and maintenance terminals;

[0008] Preferably, the cleaning robot includes a cleaning module, a driving module, a vision detection device, and an environmental sensor;

[0009] The vision detection device is used to obtain the environmental image of the photovoltaic device and judge whether there is shadow occlusion;

[0010] The environmental sensor is used to obtain the environmental parameters of the energy storage device;

[0011] The equipment management server includes a photovoltaic data module, an energy storage data module, and a robot data module;

[0012] The cloud platform server includes a regional division module, a correction and compensation module, a route planning module, and a model diagnosis module;

[0013] The regional division module is used to divide the target monitoring range containing multiple photovoltaic and energy storage subsystems into multiple target sub-regions;

[0014] The correction and compensation module is used to correct and compensate the electrical parameters of the photovoltaic equipment according to the usage parameters and environmental images of the photovoltaic equipment, and is also used to correct and compensate the electrical parameters of the energy storage equipment according to the cumulative charge and discharge times and environmental parameters of the energy storage equipment;

[0015] The route planning module is used to judge the diagnosis priority of each photovoltaic and energy storage subsystem, and plan the inspection route of the cleaning robot according to the diagnosis priority.

[0016] Preferably, the regional division module is used to obtain the position information and capacity information of each photovoltaic and energy storage subsystem within the target monitoring range;

[0017] Determine the distance between each photovoltaic and energy storage subsystem according to the position information;

[0018] Divide several photovoltaic and energy storage subsystems with a distance less than the threshold into a secondary monitoring range;

[0019] In the secondary monitoring range, according to the capacity information in each photovoltaic and energy storage subsystem, based on the capacity averaging principle, divide several photovoltaic and energy storage subsystems into a target sub-region; where the difference between the sum of the capacities of the photovoltaic and energy storage subsystems included in each target sub-region is less than a preset value.

[0020] Preferably, the route planning module is used to judge the first target subsystem according to the first diagnosis parameter and the second diagnosis parameter; the first target subsystem is a photovoltaic and energy storage subsystem with a probability of failure;

[0021] Calculate the target diagnosis parameter based on the first diagnosis parameter and the second diagnosis parameter, and judge the photovoltaic subsystem whose target diagnosis parameter exceeds the threshold as the first photovoltaic subsystem;

[0022] Judge the photovoltaic subsystem whose first diagnosis parameter exceeds the threshold as the second photovoltaic subsystem;

[0023] Judge the photovoltaic subsystem whose second diagnosis parameter exceeds the threshold as the third photovoltaic subsystem;

[0024] Set the priority of fault diagnosis: the first photovoltaic subsystem > the second photovoltaic subsystem > the third photovoltaic subsystem; the cleaning robot plans the inspection route of the cleaning robot according to the priority of fault diagnosis.

[0025] Second aspect, the present invention also provides a method for fault diagnosis of a photovoltaic and energy storage device based on a cloud platform, including:

[0026] S1. Divide the target monitoring range including multiple photovoltaic and energy storage subsystems into multiple target sub-regions;

[0027] S2. Judge the differences of the photovoltaic devices of the photovoltaic and energy storage subsystems in each target sub-region, and obtain the first diagnosis parameter;

[0028] S3. Judge the differences of the energy storage devices of the photovoltaic and energy storage subsystems in each target sub-region, and obtain the second diagnosis parameter;

[0029] S4. According to the first diagnosis parameter and the second diagnosis parameter, judge the first target subsystem; the first target subsystem is the photovoltaic and energy storage subsystem with a probability of failure;

[0030] S41. Calculate the target diagnosis parameter based on the first diagnosis parameter and the second diagnosis parameter, and judge the photovoltaic subsystem whose target diagnosis parameter exceeds the threshold as the first photovoltaic subsystem;

[0031] S42. Judge the photovoltaic subsystem whose first diagnosis parameter exceeds the threshold as the second photovoltaic subsystem;

[0032] S43. Judge the photovoltaic subsystem whose second diagnosis parameter exceeds the threshold as the third photovoltaic subsystem;

[0033] S44. Set the priority of fault diagnosis: the first photovoltaic subsystem > the second photovoltaic subsystem > the third photovoltaic subsystem;

[0034] S45. The cleaning robot plans the inspection route of the cleaning robot according to the priority of fault diagnosis, and detects the photovoltaic hot spot and the energy storage battery temperature through the visual detection device of the cleaning robot. If it is abnormal, it is judged that there is a fault, otherwise it enters S46;

[0035] S46. Take the first photovoltaic subsystem, the second photovoltaic subsystem, and the third photovoltaic subsystem as the first target subsystem;

[0036] S5. Analyze the fault cause of the cleaning robot based on the historical position and historical electrical parameters of the cleaning robot;

[0037] S6. Perform fault diagnosis on the photovoltaic and energy storage subsystems based on the fault diagnosis model.

[0038] Preferably, the S1 includes:

[0039] S11. Obtain the position information and capacity information of each photovoltaic and energy storage subsystem within the target monitoring range;

[0040] S12. Determine the distance between each photovoltaic and energy storage subsystem according to the position information;

[0041] S13. Divide several photovoltaic and energy storage subsystems with a distance less than the threshold into a secondary monitoring range;

[0042] S14. In the secondary monitoring range, based on the capacity information in each photovoltaic and energy storage subsystem and the principle of capacity averaging, divide several photovoltaic and energy storage subsystems into a target sub-region; the difference between the sum of the capacities of the photovoltaic and energy storage subsystems included in each target sub-region is less than a preset value.

[0043] Preferably, the S2 includes:

[0044] S21. Collect the electrical parameters of the photovoltaic devices in the photovoltaic and energy storage subsystems;

[0045] S22. Obtain the usage parameters of the photovoltaic devices from the device management server, including the rated power and the cumulative usage duration;

[0046] S23. Perform uniform processing on the electrical parameters based on the usage parameters of the photovoltaic devices;

[0047] U 1 =U 0 +β 1 ×T - β 2 ×P;

[0048] Wherein, U 0 is the electrical parameter of the photovoltaic device in the collected photovoltaic and energy storage subsystem, U 1 is the electrical parameter after uniform processing, T is the cumulative usage duration, P is the rated power, and β 1 , β 2 are correction factors;

[0049] S24. Obtain the environmental image of the photovoltaic device through the vision detection device of the cleaning robot, and determine whether there is shadow occlusion. If not, enter S26; if so, enter S25;

[0050] S25. Compensate the electrical parameters after uniform processing based on the shadow occlusion area;

[0051] S26. Calculate the average value of the compensated electrical parameters of the photovoltaic devices in each photovoltaic and energy storage subsystem;

[0052] S27. Calculate the difference between the compensated electrical parameters of the photovoltaic devices in each photovoltaic and energy storage subsystem and the average value, and calculate the first diagnostic parameter according to the difference.

[0053] Preferably, the S3 includes:

[0054] S31. Collect the electrical parameters of the energy storage devices in the photovoltaic and energy storage subsystems;

[0055] S32. Obtain the cumulative charge and discharge times of the energy storage device from the device management server, and correct the electrical parameters of the energy storage device;

[0056] S33. Collect the environmental parameters of the energy storage device through the environmental sensors set on the cleaning robot;

[0057] S34. Compensate the corrected electrical parameters of the energy storage device based on the environmental parameters;

[0058] S35. Calculate the average value of the compensated electrical parameters of the energy storage devices in each photovoltaic-energy storage subsystem;

[0059] S36. Calculate the difference between the compensated electrical parameters of the energy storage devices in each photovoltaic-energy storage subsystem and the average value, and calculate the second diagnostic parameter according to the difference.

[0060] Preferably, the S5 includes:

[0061] S51. Obtain the current position information of the cleaning robot, and judge whether it is in the working area. If so, enter S52; if not, enter S6;

[0062] S52. Trace the historical position information of the cleaning robot from the device management server, and the historical position information includes the historical position of the cleaning robot and the corresponding time points;

[0063] S53. Judge whether the residence time of the cleaning robot is greater than the theoretical cleaning time. If so, enter S54; otherwise, return to S51;

[0064] S54. Judge whether the current position of the cleaning robot matches the inspection route planned in S4. If so, enter S55; otherwise, judge it as a path error fault;

[0065] S55. Obtain the historical electrical parameters of the cleaning robot, and calculate whether the output power is zero. If not, enter S56; if so, judge it as a power loss fault;

[0066] S56. Extract the historical data of the drive current according to the historical electrical parameters of the cleaning robot, and judge whether there is a sudden increase in current. If so, judge it as a jamming fault; otherwise, enter step S6;

[0067] S57. Send an alarm message and the cause of the fault to the operation and maintenance terminal.

[0068] Preferably, the S6 includes:

[0069] Input the electrical parameters of the photovoltaic devices after the unified processing and compensation in S2, and the electrical parameters of the energy storage devices after the correction and compensation in S3 into the machine learning fault diagnosis model, so as to perform fault diagnosis on each first target subsystem and obtain the diagnosis result and the cause of the fault.

[0070] In a third aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method for diagnosing faults of a photovoltaic and energy storage device based on a cloud platform are implemented.

[0071] The present invention has the following beneficial effects compared with the prior art:

[0072] 1. The present invention can comprehensively diagnose and analyze faults of multiple photovoltaic and energy storage subsystems in a region. First, based on the positions and capacities of the photovoltaic and energy storage systems, multiple photovoltaic and energy storage subsystems are clustered and divided into multiple target sub-regions. By analyzing the differences, the photovoltaic and energy storage subsystems with abnormal trends are initially judged, that is, the first target subsystems. In this way, through the preliminary analysis and detection of each photovoltaic and energy storage subsystem, the computing load of the conventional fault diagnosis model can be reduced, and the efficiency and accuracy of fault diagnosis can be improved.

[0073] 2. When diagnosing the photovoltaic and energy storage system, the present invention not only uses sensors included in the photovoltaic device and the energy storage device to collect relevant parameters, but also collects the states of the photovoltaic device and the energy storage device through a cleaning robot, so as to more objectively and accurately obtain the parameter information of the photovoltaic and energy storage system and improve the accuracy of fault diagnosis.

[0074] 3. In the process of analyzing the differences of the photovoltaic device to obtain the first diagnostic parameter, the present invention performs unified processing on the electrical parameters based on the usage parameters of the photovoltaic device, and further compensates the electrical parameters based on the shadow occlusion area obtained by the visual detection device of the cleaning robot, so as to more accurately analyze the differences of the photovoltaic device. In the process of analyzing the differences of the energy storage device to obtain the second diagnostic parameter, the present invention corrects and compensates the electrical parameters based on the charge and discharge times and environmental parameters of the energy storage device, so as to more objectively and accurately analyze the differences of the energy storage device.

[0075] 4. The present invention further detects and diagnoses the first target subsystem. By collecting and analyzing the status information of the cleaning robot, it diagnoses whether the fault of the photovoltaic and energy storage system is caused by the fault of the cleaning robot, so as to quickly find the cause of the fault, enable the operation and maintenance personnel to make timely adjustments, and minimize the impact of the fault. Description of the Drawings

[0076] Figure 1 It is a schematic structural diagram of a fault diagnosis system for a photovoltaic and energy storage device based on a cloud platform provided by an embodiment of the present invention.

[0077] Figure 2 It is a flowchart of a method for diagnosing faults of a photovoltaic and energy storage device based on a cloud platform provided by an embodiment of the present invention. Detailed Embodiments

[0078] Obviously, many modifications and variations made by those skilled in the art based on the purpose of the present invention fall within the protection scope of the present invention.

[0079] Those skilled in the art can understand that unless specifically stated otherwise, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the description of the present invention means the presence of the stated features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. It should be understood that when an element or component is referred to as being "connected" to another element or component, it can be directly connected to other elements or components, or there may also be intermediate elements or components. The phrase "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0080] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0081] Embodiment 1:

[0082] The embodiment of the present invention provides a cloud platform-based fault diagnosis system for photovoltaic and energy storage devices. Specifically, please refer to Figure 1 , Figure 1 which is a schematic structural diagram of a cloud platform-based fault diagnosis system for photovoltaic and energy storage devices provided by the embodiment of the present invention. The system includes: a photovoltaic device, an energy storage device, a cleaning robot, a field monitor, a device management server, a cloud platform server, and an operation and maintenance terminal;

[0083] Preferably, the cleaning robot includes a cleaning module, a driving module, a visual detection device, and an environmental sensor;

[0084] The visual detection device is used to obtain the environmental image of the photovoltaic device and judge whether there is shadow occlusion;

[0085] The environmental sensor is used to obtain the environmental parameters of the energy storage device;

[0086] The device management server includes a photovoltaic data module, an energy storage data module, and a robot data module;

[0087] The cloud platform server includes a region division module, a correction and compensation module, a route planning module, and a model diagnosis module;

[0088] The area division module is used to divide the target monitoring range including multiple photovoltaic and energy storage subsystems into multiple target sub - regions;

[0089] The correction and compensation module is used to correct and compensate the electrical parameters of photovoltaic devices according to the usage parameters of photovoltaic devices and environmental images, and is also used to correct and compensate the electrical parameters of energy storage devices according to the cumulative charge - discharge times and environmental parameters of energy storage devices;

[0090] The route planning module is used to judge the diagnostic priorities of each photovoltaic and energy storage subsystem, and plan the inspection route of the cleaning robot according to the diagnostic priorities.

[0091] Preferably, the area division module is used to obtain the position information and capacity information of each photovoltaic and energy storage subsystem within the target monitoring range;

[0092] Determine the distances between each photovoltaic and energy storage subsystem according to the position information;

[0093] Divide several photovoltaic and energy storage subsystems with distances less than the threshold into a secondary monitoring range;

[0094] Within the secondary monitoring range, according to the capacity information in each photovoltaic and energy storage subsystem, based on the capacity averaging principle, divide several photovoltaic and energy storage subsystems into a target sub - region; where the difference between the sum of the capacities of the photovoltaic and energy storage subsystems included in each target sub - region is less than a preset value.

[0095] Preferably, the route planning module is used to judge the first target subsystem according to the first diagnostic parameter and the second diagnostic parameter; the first target subsystem is a photovoltaic and energy storage subsystem with a probability of failure;

[0096] Calculate the target diagnostic parameter based on the first diagnostic parameter and the second diagnostic parameter, and judge that the photovoltaic subsystem with the target diagnostic parameter exceeding the threshold is the first photovoltaic subsystem;

[0097] Judge that the photovoltaic subsystem with the first diagnostic parameter exceeding the threshold is the second photovoltaic subsystem;

[0098] Judge that the photovoltaic subsystem with the second diagnostic parameter exceeding the threshold is the third photovoltaic subsystem;

[0099] Set the priority of fault diagnosis: the first photovoltaic subsystem > the second photovoltaic subsystem > the third photovoltaic subsystem; the cleaning robot plans the inspection route of the cleaning robot according to the priority of fault diagnosis.

[0100] Embodiment 2:

[0101] The embodiment of the present invention also provides a method for fault diagnosis of photovoltaic and energy storage devices based on a cloud platform. For details, please refer to Figure 2 , Figure 2The flowchart of a fault diagnosis method for a photovoltaic energy storage device based on a cloud platform provided by an embodiment of the present invention. The method includes the following steps:

[0102] S1. Divide the target monitoring range including multiple photovoltaic energy storage subsystems into multiple target sub - regions;

[0103] Including:

[0104] S11. Obtain the location information and capacity information of each photovoltaic energy storage subsystem within the target monitoring range;

[0105] S12. Determine the distance between each photovoltaic energy storage subsystem according to the location information;

[0106] S13. Divide several photovoltaic energy storage subsystems with a distance less than a threshold into a secondary monitoring range;

[0107] S14. In the secondary monitoring range, based on the capacity average principle and according to the capacity information in each photovoltaic energy storage subsystem, divide several photovoltaic energy storage subsystems into a target sub - region; where the difference between the sum of the capacities of the photovoltaic energy storage subsystems included in each target sub - region is less than a preset value;

[0108] First, the environmental differences between photovoltaic energy storage subsystems with close distances are small, and the monitored parameters have certain commonalities. Therefore, they are preferably divided into one area for monitoring and diagnosis; while a photovoltaic energy storage system with a larger capacity has a larger scale, floor area, and number of devices, and the equipment, time required for monitoring and diagnosis are also larger; vice versa for a photovoltaic energy storage system with a smaller capacity; the non - uniformity of each area will lead to the non - uniformity of detection and diagnosis resources, time, and frequency within each area, and the overall efficiency will be reduced; while the present invention further divides the secondary monitoring range based on the capacity average principle to ensure that the capacities in each area are as similar as possible, thereby improving the overall detection and diagnosis efficiency.

[0109] S2. Judge the differences of the photovoltaic devices in each target sub - region of the photovoltaic energy storage subsystems, and obtain the first diagnostic parameter;

[0110] S21. Collect the electrical parameters of the photovoltaic devices of the photovoltaic energy storage subsystems;

[0111] The electrical parameters include at least one of the output voltage, output power, and output current of the photovoltaic device;

[0112] S22. Obtain the usage parameters of the photovoltaic devices from the device management server, including the rated power and the cumulative usage duration;

[0113] S23. Uniformly process the electrical parameters based on the usage parameters of the photovoltaic devices;

[0114] Among them, the electrical parameter of the photovoltaic device is selected as the output voltage U of the photovoltaic device0 Then, the output voltage U after the unified processing 1 is:

[0115] U 1 = U 0 + β 1 × T - β 2 × P;

[0116] Among them, U 0 is the electrical parameter of the photovoltaic device of the photovoltaic and energy storage subsystem collected, U 1 is the electrical parameter after the unified processing, T is the cumulative usage duration, P is the rated power, β 1 , β 2 is the correction coefficient;

[0117] S24. Obtain the environmental image of the photovoltaic device through the vision detection device of the cleaning robot, and judge whether there is shadow occlusion. If not, enter S26; if so, enter S25;

[0118] S25. Compensate the electrical parameter after the unified processing based on the shadow occlusion area;

[0119] U 2 = U 1 + β 3 × A;

[0120] Among them, is the electrical parameter after compensation based on the shadow occlusion area, A is the percentage of the shadow occlusion area, β 3 is the compensation coefficient;

[0121] Among them, the percentage of the shadow occlusion area is obtained by analyzing the photovoltaic panel image acquired by the vision detection device of the cleaning robot;

[0122] S26. Calculate the average value of the compensated electrical parameters of the photovoltaic devices of each photovoltaic and energy storage subsystem;

[0123] S27. Calculate the difference between the compensated electrical parameter of the photovoltaic device of each photovoltaic and energy storage subsystem and the average value, and calculate the first diagnosis parameter according to the difference;

[0124] Among them, the first diagnosis coefficient = the difference between the compensated electrical parameter and the average value × the first reference coefficient.

[0125] S3. Judge the differences of the energy storage devices of the photovoltaic and energy storage subsystems in each target sub-region, and obtain the second diagnosis parameter;

[0126] S31. Collect the electrical parameters of the energy storage devices of the photovoltaic and energy storage subsystems;

[0127] The electrical parameters of the energy storage device include at least one of SOC, energy storage voltage, and energy release current;

[0128] S32. Obtain the cumulative charge-discharge times of the energy storage device from the device management server, and correct the electrical parameters of the energy storage device;

[0129] E 1 = E 0 + α 1 × N;

[0130] Where, E 0 is the electrical parameter of the energy storage device of the photovoltaic-storage subsystem collected, E 1 is the corrected electrical parameter, N is the cumulative charge-discharge times, and α 1 is the correction coefficient;

[0131] S33. Collect the environmental parameters of the energy storage device through the environmental sensors set on the cleaning robot;

[0132] S34. Compensate the corrected electrical parameters of the energy storage device based on the environmental parameters;

[0133] For different types of energy storage batteries, the characteristic curves between environmental factors (such as temperature) and the electrical parameters of the energy storage device (such as SOC) are different. There are relevant literatures in the prior art. The present invention compensates the electrical parameters according to the general values of the characteristic curves, so as to improve the accuracy of diagnostic analysis.

[0134] S35. Calculate the average value of the compensated electrical parameters of the energy storage devices of each photovoltaic-storage subsystem;

[0135] S36. Calculate the difference between the compensated electrical parameters of the energy storage devices of each photovoltaic-storage subsystem and the average value, and calculate the second diagnostic parameter according to the difference;

[0136] Where, the second diagnostic coefficient = the difference between the compensated electrical parameter and the average value × the second reference coefficient.

[0137] S4. Determine the first target subsystem according to the first diagnostic parameter and the second diagnostic parameter; the first target subsystem is the photovoltaic-storage subsystem with a failure probability;

[0138] S41. Calculate the target diagnostic parameter based on the first diagnostic parameter and the second diagnostic parameter, and determine the photovoltaic subsystem for which the target diagnostic parameter exceeds the threshold as the first photovoltaic subsystem;

[0139] Where, the target diagnostic parameter = γ 1 × the first diagnostic parameter + γ 2 × the second diagnostic parameter; γ 1 、γ 2 are adjustable weight coefficients;

[0140] S42. Determine the photovoltaic subsystem for which the first diagnostic parameter exceeds the threshold as the second photovoltaic subsystem;

[0141] S43, determining that the photovoltaic subsystem whose second diagnostic parameter exceeds the threshold is the third photovoltaic subsystem;

[0142] S44, setting the priority of fault diagnosis: the first photovoltaic subsystem> the second photovoltaic subsystem> the third photovoltaic subsystem;

[0143] S45, the cleaning robot plans the inspection route of the cleaning robot according to the priority of fault diagnosis, and detects the photovoltaic hot spot and the temperature of the energy storage battery through the detection device of the cleaning robot. If abnormal, it is determined that there is a fault, otherwise it goes to S46;

[0144] The cleaning robot itself is designed with a cleaning route. In step S45, the positions of the first photovoltaic subsystem, the second photovoltaic subsystem, and the third photovoltaic subsystem are added to the cleaning route, and the path is replanned to form an inspection route.

[0145] S46, taking the first photovoltaic subsystem, the second photovoltaic subsystem, and the third photovoltaic subsystem as first target subsystems;

[0146] S5: Analyze the cause of the cleaning robot's failure based on the cleaning robot's historical position and historical electrical parameters;

[0147] S51, obtaining the current position information of the cleaning robot, determining whether it is in the working area, if so, proceeding to S52, if not, proceeding to S6;

[0148] If the cleaning robot is not in the working area, there is no occlusion fault, so the process enters S6 and uses the diagnostic model to perform fault diagnosis.

[0149] S52, tracing back historical location information of the cleaning robot from the device management server, wherein the historical location information includes historical locations of the cleaning robot and corresponding time points;

[0150] S53, judging whether the residence time of the cleaning robot is greater than the theoretical cleaning time, if so, proceeding to S54, otherwise returning to S51;

[0151] S54, determining whether the current position of the cleaning robot matches the inspection route planned in S4, if so, proceeding to S55, otherwise determining that it is a path error fault;

[0152] S55, obtaining the historical electrical parameters of the cleaning robot, calculating whether the output power is zero, if not, proceeding to S56, if yes, determining that it is a power failure;

[0153] S56, extracting the historical data of the driving current according to the historical electrical parameters of the cleaning robot, and determining whether a sudden increase in current occurs, if so, determining that it is a stuck fault, otherwise proceeding to step S6;

[0154] Among them, if the above-mentioned faults do not occur, it enters S6 to perform fault diagnosis using the diagnostic model.

[0155] S57. Send the alarm information and the cause of the fault to the operation and maintenance terminal;

[0156] Among them, after receiving the alarm information and the cause of the fault, the operation and maintenance personnel notify the on-site personnel of the photovoltaic and energy storage subsystem to eliminate the fault and then restart the diagnosis from S51;

[0157] S6. Perform fault diagnosis on the photovoltaic and energy storage subsystem based on the fault diagnosis model;

[0158] Including: inputting the electrical parameters of the photovoltaic equipment after unified processing and compensation in S2, and the electrical parameters of the energy storage equipment after correction and compensation in S3 into the machine learning fault diagnosis model, so as to perform fault diagnosis on each first target subsystem and obtain the diagnosis result and the cause of the fault.

[0159] The faults include: hot spot fault, energy storage fault, energy release fault, standby fault, etc.

[0160] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. 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 description of the method part.

[0161] Professionals can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in the form of hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0162] Finally, it should also be noted that in this article, relationships such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "including", "comprising" or any other variant is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

Claims

1. A cloud platform-based photovoltaic storage equipment fault diagnosis system, including photovoltaic equipment, energy storage equipment, cleaning robots, on-site monitoring machines, equipment management servers, cloud platform servers, and operation and maintenance terminals; characterized in that: The cleaning robot includes a cleaning module, a driving module, a visual detection device, and an environmental sensor; the visual detection device is used to obtain the environmental image of the photovoltaic device and determine whether there is a shadow; the environmental sensor is used to obtain the environmental parameters of the energy storage device; the device management server includes a photovoltaic data module, an energy storage data module, and a robot data module; the cloud platform server includes an area division module, a correction and compensation module, a route planning module, and a model diagnosis module; the area division module is used to divide the target monitoring range containing multiple photovoltaic storage subsystems into multiple target sub-areas; the correction and compensation module is used to correct and compensate the electrical parameters of the photovoltaic device according to the use parameters and environmental parameters of the photovoltaic device, and is also used to correct and compensate the electrical parameters of the energy storage device according to the cumulative charge and discharge times and environmental parameters of the energy storage device; the route planning module is used to determine the diagnostic priority of each photovoltaic storage subsystem, and plan the inspection route of the cleaning robot according to the diagnostic priority.

2. The optical storage device fault diagnosis system based on cloud platform according to claim 1 is characterized in that: The area division module is used to obtain the location information and capacity information of each photovoltaic storage subsystem within the target monitoring range; determine the distance between each photovoltaic storage subsystem based on the location information; divide a number of photovoltaic storage subsystems whose distance is less than a threshold into a secondary monitoring range; in the secondary monitoring range, according to the capacity information in each photovoltaic storage subsystem, based on the capacity averaging principle, divide the several photovoltaic storage subsystems into a target sub-area; wherein the difference between the sum of the capacities of the photovoltaic storage subsystems included in each target sub-area is less than a preset value.

3. The optical storage device fault diagnosis system based on cloud platform according to claim 2 is characterized in that: The route planning module is used to determine the first target subsystem according to the first diagnostic parameter and the second diagnostic parameter; the first target subsystem is a photovoltaic storage subsystem with a probability of failure; the target diagnostic parameter is calculated based on the first diagnostic parameter and the second diagnostic parameter, and the photovoltaic subsystem whose target diagnostic parameter exceeds the threshold is determined to be the first photovoltaic subsystem; the photovoltaic subsystem whose first diagnostic parameter exceeds the threshold is determined to be the second photovoltaic subsystem; the photovoltaic subsystem whose second diagnostic parameter exceeds the threshold is determined to be the third photovoltaic subsystem; the priority of fault diagnosis is set: first photovoltaic subsystem>second photovoltaic subsystem>third photovoltaic subsystem; and the inspection route of the cleaning robot is planned according to the priority of fault diagnosis.

4. A cloud platform-based optical storage device fault diagnosis method applied to the system according to any one of claims 1 to 3, characterized in that: The method includes: S1. Divide a target monitoring range including multiple photovoltaic storage subsystems into multiple target sub-areas; S2, determining the difference of photovoltaic devices of the photovoltaic storage subsystem in each target sub-area, and obtaining a first diagnostic parameter; S3, determining the difference of energy storage devices of the photovoltaic storage subsystem in each target sub-area, and obtaining a second diagnostic parameter; S4. Determine a first target subsystem according to the first diagnostic parameter and the second diagnostic parameter; the first target subsystem is a photovoltaic storage subsystem with a probability of failure; S41, calculating a target diagnostic parameter based on the first diagnostic parameter and the second diagnostic parameter, and determining that the photovoltaic subsystem whose target diagnostic parameter exceeds a threshold is the first photovoltaic subsystem; S42, determining that the photovoltaic subsystem whose first diagnostic parameter exceeds the threshold is the second photovoltaic subsystem; S43, determining that the photovoltaic subsystem whose second diagnostic parameter exceeds the threshold is the third photovoltaic subsystem; S44, setting the priority of fault diagnosis: the first photovoltaic subsystem> the second photovoltaic subsystem> the third photovoltaic subsystem; S45, the cleaning robot plans the inspection route of the cleaning robot according to the priority of fault diagnosis, and detects photovoltaic hot spots and energy storage battery temperature through the visual detection device of the cleaning robot. If abnormal, it is determined that there is a fault, otherwise it goes to S46; S46, taking the first photovoltaic subsystem, the second photovoltaic subsystem, and the third photovoltaic subsystem as first target subsystems; S5. Analyze the cause of the cleaning robot's failure based on the cleaning robot's historical position and historical electrical parameters; S6. Perform fault diagnosis on the optical storage subsystem based on the fault diagnosis model.

5. The optical storage device fault diagnosis method based on cloud platform according to claim 4 is characterized in that: The S1 includes: S11, obtaining location information and capacity information of each photovoltaic storage subsystem within the target monitoring range; S12, determining the distance between each optical storage subsystem according to the position information; S13, dividing a number of photovoltaic storage subsystems whose distances are less than a threshold into a secondary monitoring range; S14. In the secondary monitoring range, according to the capacity information in each photovoltaic storage subsystem and based on the capacity averaging principle, several photovoltaic storage subsystems are divided into a target sub-area; wherein the difference between the sum of the capacities of the photovoltaic storage subsystems included in each target sub-area is less than a preset value.

6. The optical storage device fault diagnosis method based on cloud platform according to claim 5 is characterized in that: The S2 includes: S21, collecting electrical parameters of the photovoltaic equipment of the light storage subsystem; S22, obtaining the usage parameters of the photovoltaic equipment from the equipment management server, including the rated power and the accumulated usage time; S23, performing unified processing on the electrical parameters based on the usage parameters of the photovoltaic equipment; U1=U0+β1×T-β2×P; Among them, U0 is the collected electrical parameters of the photovoltaic equipment of the photovoltaic storage subsystem, U1 is the electrical parameters after unified processing, T is the cumulative usage time, P is the rated power, β1 and β2 are correction coefficients; S24, obtaining environmental parameters of the photovoltaic equipment through the visual detection device of the cleaning robot, and determining whether there is shadow blocking, if not, proceeding to S26, if yes, proceeding to S25; S25, compensating the electrical parameters after the unified processing based on the shadow shielding area; S26, calculating the average value of the compensated electrical parameters of the photovoltaic devices of each photovoltaic storage subsystem; S27, calculating the difference between the compensated electrical parameters of the photovoltaic devices of each photovoltaic storage subsystem and the average value, and calculating the first diagnostic parameter according to the difference.

7. The optical storage device fault diagnosis method based on cloud platform according to claim 6 is characterized in that: The S3 includes: S31, collecting electrical parameters of the energy storage device of the photovoltaic storage subsystem; S32, obtaining the cumulative number of charge and discharge times of the energy storage device from the device management server, and correcting the electrical parameters of the energy storage device; S33, collecting environmental parameters of the energy storage device through an environmental sensor provided on the cleaning robot; S34, compensating the corrected electrical parameters of the energy storage device based on the environmental parameters; S35, calculating the average value of the compensated electrical parameters of the energy storage devices of each photovoltaic storage subsystem; S36, calculating the difference between the compensated electrical parameter of the energy storage device of each photovoltaic storage subsystem and the average value, and calculating the second diagnostic parameter according to the difference.

8. The optical storage device fault diagnosis method based on cloud platform according to claim 7 is characterized in that: The S5 includes: S51, obtaining the current position information of the cleaning robot, determining whether it is in the working area, if so, proceeding to S52, if not, proceeding to S6; S52, tracing back historical location information of the cleaning robot from the device management server, wherein the historical location information includes historical locations of the cleaning robot and corresponding time points; S53, judging whether the residence time of the cleaning robot is greater than the theoretical cleaning time, if so, proceeding to S54, otherwise returning to S51; S54, determining whether the current position of the cleaning robot matches the inspection route planned in S4, if so, proceeding to S55, otherwise determining that it is a path error fault; S55, obtaining the historical electrical parameters of the cleaning robot, calculating whether the output power is zero, if not, proceeding to S56, if yes, determining that it is a power failure; S56, extracting the historical data of the driving current according to the historical electrical parameters of the cleaning robot, and determining whether a sudden increase in current occurs, if so, determining that it is a stuck fault, otherwise proceeding to step S6; S57. Send alarm information and fault cause to the operation and maintenance terminal.

9. The optical storage device fault diagnosis method based on cloud platform according to claim 8 is characterized in that: The S6 includes: The electrical parameters of the photovoltaic equipment after unified processing and compensation in S2, and the electrical parameters of the energy storage equipment after correction and compensation in S3, are input into the machine learning fault diagnosis model, so as to perform fault diagnosis on each first target subsystem and obtain the diagnosis result and the cause of the fault.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cloud platform-based optical storage device fault diagnosis method described in any one of claims 4 to 8 are implemented.

Images