Photovoltaic power station fault diagnosis method and device and storage medium

Abstract

The invention relates to the field of photovoltaic power stations, and discloses a photovoltaic power station fault diagnosis method and device and a storage medium. The photovoltaic power station fault diagnosis method comprises the steps that multi-modal data and a fault diagnosis model of a photovoltaic power station area are acquired, the multi-modal data comprise image data and time sequence data, and the fault diagnosis model comprises a convolutional neural network model and a long-short term memory network model; and inputting the multi-modal data into the fault diagnosis model, and generating a fault detection result of the photovoltaic power station area. According to the photovoltaic power station fault diagnosis method provided by the embodiment of the invention, comprehensive and accurate fault detection can be carried out on the photovoltaic power station area according to the multi-modal data.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power stations, and in particular to a photovoltaic power station fault diagnosis method, equipment and storage medium. Background Art

[0002] With the development of renewable energy technology, the number of photovoltaic power stations is increasing. As a clean energy power generation facility, the stability and power generation efficiency of photovoltaic power stations during operation are important indicators. When a photovoltaic power station is in operation, it is easy to cause damage to the equipment of the photovoltaic power station due to its own faults or external damage, which in turn affects the stability and power generation efficiency during operation.

[0003] Traditional fault detection methods rely on manual inspections or simple alarm systems based on fixed thresholds. These methods often fail to detect early signs of faults in PV power plants in a timely manner, nor can they effectively distinguish fault types and predict fault development trends.

[0004] For example, the existing technology generally only detects a single type of data through sensors, such as detecting the current and voltage change data of the photovoltaic power station through current and voltage sensors, or relying solely on visual sensors to detect the surface image data of the photovoltaic power station equipment. Different types of data are isolated from each other, so this simple detection method is difficult to accurately detect faults in photovoltaic power stations. Summary of the invention

[0005] The embodiments of the present invention provide a photovoltaic power station fault diagnosis method, device and storage medium to solve the above problems.

[0006] In a first aspect, an embodiment of the present invention provides a photovoltaic power station fault diagnosis method, which includes: acquiring multimodal data and a fault diagnosis model of a photovoltaic power station area, the multimodal data including image data and time series data, the fault diagnosis model including a convolutional neural network model and a long short-term memory network model; inputting the multimodal data into the fault diagnosis model to generate a fault detection result of the photovoltaic power station area.

[0007] The photovoltaic power station fault diagnosis method provided in the embodiment of the present invention can integrate different data in the photovoltaic power station area to obtain multimodal data and a fault diagnosis model. After the multimodal data is input into the fault diagnosis model, comprehensive fault detection can be achieved in the photovoltaic power station area to improve the accuracy of fault detection in the photovoltaic power station.

[0008] Optionally, the photovoltaic power station area is divided into multiple collection sub-areas, each collection sub-area includes at least one photovoltaic device, and the step of obtaining multimodal data of the photovoltaic power station area includes: obtaining multimodal data of the photovoltaic devices in each collection sub-area; based on a data fusion algorithm, processing the multimodal data in each collection sub-area to obtain multimodal data of the entire photovoltaic power station area.

[0009] Optionally, the step of obtaining the fault diagnosis model includes: using image data as input to a convolutional neural network model so that the convolutional neural network model outputs an image feature vector; using time series data as input to a long short-term memory network model so that the long short-term memory network model outputs a time series vector.

[0010] Optionally, the step of obtaining the fault diagnosis model further includes: connecting the image feature vector and the time series vector to form a fault detection vector; and connecting the fault detection vector and the fully connected layer to form a fault diagnosis model.

[0011] Optionally, the step of obtaining the fault diagnosis model includes: using the image data as the input of the convolutional neural network model, so that the convolutional neural network model outputs an image feature vector; using the image feature vector and time series data as the input of the long short-term memory network model, so that the long short-term memory network model outputs a fault detection vector.

[0012] Optionally, the step of inputting multimodal data into a fault diagnosis model to generate fault detection results for the photovoltaic power station area includes: obtaining a preset acquisition frequency of the fault diagnosis model; based on the preset acquisition frequency, collecting multimodal data of the photovoltaic power station area; inputting the multimodal data into the fault diagnosis model to generate fault detection results.

[0013] Optionally, after the step of inputting the multimodal data into the fault diagnosis model to generate the fault detection results of the photovoltaic power station area, the method also includes: calculating the probability of fault occurrence based on the fault detection results; generating a warning message of a corresponding level according to the magnitude of the fault probability; and generating fault area information of the photovoltaic power station area based on the fault detection results.

[0014] Optionally, the image data includes appearance images of photovoltaic equipment in the photovoltaic power station area, and the time series data includes temperature information, humidity information, current information, voltage information of the photovoltaic equipment and meteorological information in the photovoltaic power station area.

[0015] In a second aspect, an embodiment of the present invention provides a photovoltaic power station fault diagnosis device, comprising: a processor and a memory, wherein instructions are stored in the memory; the processor calls the instructions in the memory so that the processor executes the photovoltaic power station fault diagnosis method of any of the aforementioned embodiments of the first aspect of the present invention.

[0016] The processor of the photovoltaic power station fault diagnosis device provided by the embodiment of the present invention executes the photovoltaic power station fault diagnosis method of any of the aforementioned embodiments of the first aspect of the present invention by calling the instructions in the memory, and can integrate different data in the photovoltaic power station area to obtain multimodal data, and input the multimodal data into the fault diagnosis model to generate fault detection results, so that the fault diagnosis model can realize comprehensive fault detection of the photovoltaic power station area, so as to improve the accuracy of fault detection of the photovoltaic power station.

[0017] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed by a processor, a photovoltaic power station fault diagnosis method according to any of the aforementioned embodiments of the first aspect of the present invention is implemented.

[0018] The instructions stored in the computer-readable storage medium provided in the embodiment of the present invention can be called by the processor and execute the photovoltaic power station fault diagnosis method of any of the aforementioned embodiments of the first aspect of the present invention, so as to integrate different data in the photovoltaic power station area to obtain multimodal data, and input the multimodal data into the fault diagnosis model to generate fault detection results, so that the fault diagnosis model can realize comprehensive fault detection of the photovoltaic power station area, so as to improve the accuracy of fault detection of the photovoltaic power station. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] 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 the structures shown in these drawings without paying creative work.

[0020] Figure 1 This is a flow chart of a first embodiment of a photovoltaic power station fault diagnosis method according to the present invention;

[0021] Figure 2 This is a flow chart of obtaining multi-modal data in step S110 of the first embodiment of the photovoltaic power station fault diagnosis method of the present invention;

[0022] Figure 3 This is a flow chart of the case where a parallel combination method is used in step S110 of the first embodiment of the photovoltaic power station fault diagnosis method of the present invention;

[0023] Figure 4 This is a flow chart of the case where a cascade combination mode is adopted in step S110 of the first embodiment of the photovoltaic power station fault diagnosis method of the present invention;

[0024] Figure 5This is a flow chart of step S120 of the first embodiment of the photovoltaic power station fault diagnosis method of the present invention;

[0025] Figure 6 is a flow chart of a second embodiment of a photovoltaic power station fault diagnosis method according to the present invention;

[0026] Figure 7 A structural block diagram of an embodiment of a photovoltaic power station fault diagnosis device of the present invention;

[0027] Figure 8 A schematic diagram of the structure of an embodiment of a photovoltaic power station fault diagnosis device of the present invention;

[0028] Fig. 9 The figure is a structural block diagram of an embodiment of a photovoltaic power station fault diagnosis device of the present invention. DETAILED DESCRIPTION

[0029] 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.

[0030] It should be noted that all directional indications such as up, down, left, right, front, back, etc. in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture as shown in the accompanying drawings. If the specific posture changes, the directional indication will also change accordingly.

[0031] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0032] The embodiment of the present invention proposes a photovoltaic power station fault diagnosis method, which is used to monitor the operating status of photovoltaic equipment in the photovoltaic power station area and generate fault detection results when the photovoltaic equipment fails, so that maintenance personnel can promptly find the cause of the fault and repair the photovoltaic equipment.

[0033] For ease of understanding, the photovoltaic power station fault diagnosis method of the first embodiment of the present invention is described below. Figures 1 to 4 As shown, the photovoltaic power station fault diagnosis method in the first embodiment of the present invention includes steps S110 to S120.

[0034] In step S110, multimodal data and a fault diagnosis model of a photovoltaic power station area are obtained, wherein the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model.

[0035] In some optional embodiments, the photovoltaic power station area is divided into a plurality of collection sub-areas, and each collection sub-area includes at least one photovoltaic device.

[0036] like Figure 2 As shown, in step S110, the step of acquiring multimodal data of the photovoltaic power station area includes steps S111 to S112.

[0037] In step S111 , multimodal data of photovoltaic equipment in each acquisition sub-area is acquired.

[0038] In step S112, based on the data fusion algorithm, the multimodal data in each acquisition sub-area is processed to obtain the multimodal data of the entire photovoltaic power station area.

[0039] In this embodiment, the photovoltaic power station area is divided into multiple collection sub-areas to obtain multimodal data of the photovoltaic equipment in each collection sub-area, and the multimodal data of the photovoltaic equipment in all collection sub-areas are integrated through a data fusion algorithm to reduce interference in the multimodal data and improve the accuracy of the multimodal data, thereby obtaining the multimodal data of the entire photovoltaic power station area.

[0040] Among them, this embodiment can use data fusion algorithms such as Kalman filtering and particle filtering to integrate the multimodal data of photovoltaic devices in each collection sub-area.

[0041] In some optional embodiments, the image data includes appearance images of photovoltaic equipment in the photovoltaic power station area, and the time series data includes temperature information, humidity information, current information, voltage information of the photovoltaic equipment and meteorological information in the photovoltaic power station area.

[0042] In this embodiment, the appearance image of the photovoltaic device can be used to determine whether the surface of the photovoltaic device has cracks, dirt, dust, electrical connection failure, etc. that affect the normal operation of the photovoltaic device. The meteorological information includes wind direction information and wind speed information in the photovoltaic power station area.

[0043] like Figure 3As shown, in some optional embodiments, the step of acquiring the fault diagnosis model includes steps S113 to S116.

[0044] In step S113, the image data is used as the input of the convolutional neural network model, so that the convolutional neural network model outputs an image feature vector.

[0045] In step S114, the time series data is used as the input of the long short-term memory network model, so that the long short-term memory network model outputs a time series vector.

[0046] In this embodiment, the convolutional neural network model and the long short-term memory network model are combined in parallel, that is, the convolutional neural network model and the long short-term memory network model output vectors respectively, and the output vectors of the two are connected together to obtain a fault diagnosis model. The specific steps of the parallel combination method are as follows:

[0047] In step S115, the image feature vector is connected with the time series vector to form a fault detection vector.

[0048] In step S116, the fault detection vector is connected to the fully connected layer to form a fault diagnosis model.

[0049] In this embodiment, the image feature vector output by the convolutional neural network model and the time series vector output by the long short-term memory network model are connected to form a fault detection vector, and the fault detection vector is connected to the fully connected layer so that the fully connected layer can process the output of the convolutional neural network model and the output of the long short-term memory network model to form a fault diagnosis model.

[0050] like Figure 4 As shown, in some other optional embodiments, the step of acquiring the fault diagnosis model includes steps S117 to S118.

[0051] In step S117, the image data is used as the input of the convolutional neural network model, so that the convolutional neural network model outputs an image feature vector.

[0052] In step S118, the image feature vector and the time series data are used as inputs of the long short-term memory network model, so that the long short-term memory network model outputs a fault detection vector.

[0053] In this embodiment, the convolutional neural network model and the long short-term memory network model are combined in a cascade manner, that is, the image feature vector output by the convolutional neural network model is used as the input vector of the long short-term memory network model, so that the image feature vector and the time series data are used as the input vector of the long short-term memory network model together, and then the long short-term memory network model outputs a fault detection vector to form a fault diagnosis model.

[0054] Among them, since the image feature vector output by the convolutional neural network model is used as the input vector of the long short-term memory network model, the fault detection vector output by the long short-term memory network model includes the image feature vector and the time series vector, which is used for fault detection of photovoltaic equipment in the photovoltaic power station area.

[0055] Specifically, the convolutional neural network model and the long short-term memory network model are constructed and obtained through the following steps:

[0056] For image data, the acquired appearance image can be first scaled to a uniform size, such as 224x224 pixels, and then the appearance image can be normalized so that its value is between 0 and 1. For time series data, smoothing processing can be performed first to remove data noise and realize normalization processing so that its value is between 0 and 1.

[0057] An initial convolutional neural network model and an initial long short-term memory network model are established respectively. The initial convolutional neural network model includes a convolutional layer, a pooling layer, and a fully connected layer. The initial convolutional neural network model is trained with image data to obtain a convolutional neural network model to be acquired, which is used to identify features in the appearance image, such as whether there are cracks, dust, etc. on the component surface, and finally output an image feature vector.

[0058] The initial LSTM network model is trained by time series data to obtain the LSTM network model to be acquired, which is used to process time series data, such as temperature information, humidity information, current information, voltage information of photovoltaic equipment and meteorological information in the photovoltaic power station area, and finally output a time series vector.

[0059] In step S120, the multimodal data is input into a fault diagnosis model to generate fault detection results for the photovoltaic power station area.

[0060] like Figure 5 As shown, in some optional embodiments, step S130 includes steps S131 to S133.

[0061] In step S121 , a preset acquisition frequency of the fault diagnosis model is obtained.

[0062] In step S122, multimodal data of the photovoltaic power station area is collected based on a preset collection frequency.

[0063] In step S123, the multimodal data is input into the fault diagnosis model to generate a fault detection result.

[0064] In this embodiment, multimodal data of the photovoltaic power station area is collected at a preset collection frequency. After a period of collection, the operating status of the photovoltaic equipment can be obtained based on the changes in the image data. The change trend graph of the voltage, current, temperature, humidity and meteorological information of the photovoltaic equipment can also be formed based on the changes in the time series data, so as to generate fault detection results.

[0065] Specifically, the preset acquisition frequencies of the image data and the time series data may be the same or different. For example, the image data may be preset to capture the appearance image of the photovoltaic device once every 24 hours, and the time series data may be preset to capture the voltage, current, temperature, humidity of the photovoltaic device and the meteorological information of the photovoltaic power station area once every 15 minutes.

[0066] Among them, image data can be obtained by taking appearance images of photovoltaic equipment at fixed time and fixed area through cameras, camera components or drones equipped with camera components, and time series data can be obtained through current sensors, voltage sensors, temperature sensors, humidity sensors and weather stations set up in the photovoltaic power station area.

[0067] like Figure 6 As shown, in the second embodiment of the present invention, the photovoltaic power station fault diagnosis method includes steps S210 to S240.

[0068] Some steps of the second embodiment are the same as those of the first embodiment. The differences between the two embodiments will be described below, and the similarities will not be described in detail again.

[0069] In the second embodiment of the present invention, step S210 to step S220 are the same as those in the first embodiment and will not be described in detail herein.

[0070] In step S210, multimodal data and a fault diagnosis model of a photovoltaic power station area are obtained. The multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model.

[0071] In step S220, the multimodal data is input into a fault diagnosis model to generate fault detection results for the photovoltaic power station area.

[0072] In step S230 , the probability of a fault occurring is calculated based on the fault detection result.

[0073] In step S240, a warning message of a corresponding level is generated according to the magnitude of the probability of the failure occurring.

[0074] In some optional embodiments, step S240 further includes: generating fault area information of the photovoltaic power station area based on the fault detection result.

[0075] In this embodiment, the fault diagnosis model can generate fault detection results based on the multimodal data of the photovoltaic power station area, and then calculate the failure probability of photovoltaic equipment in the photovoltaic power station area. For example, during the operation of the photovoltaic power station, possible faults include: local overheating, photovoltaic equipment surface being blocked, cracks, dust, electrical connection failure or surface aging.

[0076] The reasons for local overheating include that when part of the photovoltaic equipment is blocked or shadowed, the area cannot generate enough electricity, but instead becomes a load, consuming the electricity generated by other parts, causing local temperature rise and local overheating. The reasons for cracks include physical damage, bad weather (such as hail, strong winds), material aging, etc., which cause cracks on the surface of photovoltaic equipment. The reason for dust is that dust in the air, fallen leaves, etc. accumulate on the surface of photovoltaic equipment. The reason for electrical connection failure is that the wiring of photovoltaic equipment is loose, corroded or has poor contact, resulting in unstable current transmission of photovoltaic equipment. The reason for aging is that photovoltaic equipment is exposed to outdoor environment for a long time, causing the surface material of photovoltaic equipment to gradually age and performance to decline.

[0077] After discovering the cause and location of the fault, maintenance personnel can perform repairs according to the different fault types of the photovoltaic equipment. For example, maintenance personnel can remove obstructions, set bypass diodes on the photovoltaic equipment to avoid faulty photovoltaic equipment, directly replace damaged photovoltaic equipment, replace the outer casing of the photovoltaic equipment, regularly clean the photovoltaic equipment, regularly check the connection lines of the photovoltaic equipment, install protective nets, etc., to solve the fault of the photovoltaic equipment and ensure the normal operation of the photovoltaic power station.

[0078] The fault detection results obtained by the method of this application calculate the probability of fault occurrence, and the probability of local overheating fault occurrence is 80%, the probability of shading occurrence is 20%, and the probability of crack occurrence is 10%. The higher the fault probability, the greater the possibility of the photovoltaic equipment having the fault. Maintenance personnel can give priority to solving faults with higher probability of occurrence according to the level of fault probability to improve maintenance efficiency.

[0079] In this embodiment, the greater the probability of a fault, the higher the level of the warning information. For example, when the probability of a fault reaches 20%, a first level warning is issued, when the probability of a fault reaches 50%, a second level warning is issued, and when the probability of a fault reaches 80%, a third level warning is issued. The higher the warning level, the higher the priority of resolving the fault. The fault area information includes the area where the faulty photovoltaic equipment is located in the photovoltaic power station area, so that maintenance personnel can quickly determine what kind of fault the photovoltaic equipment in which area of ​​the photovoltaic power station area has occurred based on the warning level and fault area information, so as to improve maintenance efficiency.

[0080] The photovoltaic power station fault diagnosis method provided by the embodiment of the present invention includes: acquiring multimodal data and a fault diagnosis model of the photovoltaic power station area, the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model; inputting the multimodal data into the fault diagnosis model to generate a fault detection result of the photovoltaic power station area.

[0081] The photovoltaic power station fault diagnosis method provided by the embodiment of the present invention can integrate different data in the photovoltaic power station area to obtain multimodal data. After the multimodal data is input into the fault diagnosis model, comprehensive fault detection can be achieved in the photovoltaic power station area to improve the accuracy of fault detection in the photovoltaic power station.

[0082] For the above method embodiment, the embodiment of the present invention also provides Figure 7 A photovoltaic power station fault diagnosis device shown includes an acquisition module 210 and a detection module 220 .

[0083] The acquisition module 210 is used to acquire multimodal data and a fault diagnosis model of the photovoltaic power station area, wherein the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model.

[0084] The detection module 220 is used to input the multimodal data into the fault diagnosis model to generate fault detection results for the photovoltaic power station area.

[0085] In this embodiment, the acquisition module 210 includes a plurality of acquisition units 211 , and the acquisition units 211 are used to monitor the operating status of photovoltaic devices in the photovoltaic power station area and acquire multi-modal data.

[0086] like Figure 8 As shown, the photovoltaic power station area is divided into multiple areas, such as the first area R1, the second area R2, and up to the Nth area RN. Each area is equipped with a collection unit 211 to collect multimodal data of photovoltaic equipment in the area. After the fault diagnosis model is obtained by the acquisition module 210, the fault detection result of the photovoltaic power station area is generated by the detection module 220 to achieve comprehensive fault detection of the photovoltaic power station area. At the same time, it can also accurately warn the area where the fault occurs in the photovoltaic power station area, so that maintenance personnel can find faulty equipment in time, shorten the downtime of the photovoltaic power station, and speed up the maintenance.

[0087] The photovoltaic power station fault diagnosis device provided by the embodiment of the present invention can integrate different data in the photovoltaic power station area to obtain multimodal data by implementing the above method, and input the multimodal data into the fault diagnosis model to generate fault detection results, so that the fault diagnosis model can realize comprehensive fault detection in the photovoltaic power station area, so as to improve the accuracy of fault detection in the photovoltaic power station.

[0088] For the above method embodiment, the embodiment of the present invention also provides Fig. 9 A photovoltaic power station fault diagnosis device shown includes: a processor 301 and a memory 302, and the memory 302 stores instructions.

[0089] The processor 301 can call the instructions in the memory 302 so that the processor 301 executes the photovoltaic power station fault diagnosis method according to any of the aforementioned embodiments of the present invention.

[0090] The photovoltaic power station fault diagnosis method includes: obtaining multimodal data and a fault diagnosis model of a photovoltaic power station area, the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model; inputting the multimodal data into the fault diagnosis model to generate a fault detection result of the photovoltaic power station area.

[0091] The processor 301 of the photovoltaic power station fault diagnosis device provided in the embodiment of the present invention executes the photovoltaic power station fault diagnosis method of any of the aforementioned embodiments of the present invention by calling the instructions in the memory 302, and can integrate different data in the photovoltaic power station area to obtain multimodal data and a fault diagnosis model. After the multimodal data is input into the fault diagnosis model, comprehensive fault detection can be achieved in the photovoltaic power station area to improve the accuracy of fault detection of the photovoltaic power station.

[0092] Furthermore, the photovoltaic power station fault diagnosis device provided by the embodiment of the present invention may further include a communication interface 303 and a bus 304 , and the processor 301 , the memory 302 and the communication interface 303 are electrically connected via the bus 304 .

[0093] The memory 302 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 303 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 304 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig. 9Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0094] The processor 301 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 301. The above processor 301 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 302, and the processor 301 reads the information in the memory 302 and completes the steps of the method of the above embodiment in combination with its hardware.

[0095] An embodiment of the present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the above-mentioned photovoltaic power station fault diagnosis method.

[0096] The photovoltaic power station fault diagnosis method includes: obtaining multimodal data and a fault diagnosis model of a photovoltaic power station area, the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model; inputting the multimodal data into the fault diagnosis model to generate a fault detection result of the photovoltaic power station area.

[0097] The instructions stored in the computer-readable storage medium provided in the embodiment of the present invention can integrate different data of the photovoltaic power station area to obtain multimodal data, and input the multimodal data into the fault diagnosis model to generate fault detection results, so that the fault diagnosis model can realize comprehensive fault detection of the photovoltaic power station area, so as to improve the accuracy of fault detection of the photovoltaic power station.

[0098] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0099] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0100] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic power station fault diagnosis method, characterized in that: The method comprises: Acquire multimodal data and a fault diagnosis model of a photovoltaic power station area, wherein the multimodal data includes image data and time series data, and the fault diagnosis model includes a convolutional neural network model and a long short-term memory network model; The multimodal data is input into the fault diagnosis model to generate a fault detection result of the photovoltaic power station area.

2. The photovoltaic power station fault diagnosis method according to claim 1, characterized in that: The photovoltaic power station area is divided into a plurality of acquisition sub-areas, each of which includes at least one photovoltaic device. The step of acquiring multimodal data of the photovoltaic power station area includes: Acquiring the multimodal data of the photovoltaic device in each of the acquisition sub-areas; Based on a data fusion algorithm, the multimodal data in each of the acquisition sub-areas is processed to obtain the multimodal data of the entire photovoltaic power station area.

3. The photovoltaic power station fault diagnosis method according to claim 1, characterized in that: The steps to obtain the fault diagnosis model include: Using the image data as input of the convolutional neural network model, so that the convolutional neural network model outputs an image feature vector; The time series data is used as the input of the long short-term memory network model, so that the long short-term memory network model outputs a time series vector.

4. The photovoltaic power station fault diagnosis method according to claim 3, characterized in that: The steps of obtaining the fault diagnosis model also include: Connecting the image feature vector and the time series vector to form a fault detection vector; The fault detection vector is connected to a fully connected layer to form the fault diagnosis model.

5. The photovoltaic power station fault diagnosis method according to claim 1, characterized in that: The steps to obtain the fault diagnosis model include: Using the image data as input of the convolutional neural network model, so that the convolutional neural network model outputs an image feature vector; The image feature vector and the time series data are used as inputs of the long short-term memory network model, so that the long short-term memory network model outputs a fault detection vector.

6. The photovoltaic power station fault diagnosis method according to claim 1, characterized in that: The step of inputting the multimodal data into the fault diagnosis model to generate the fault detection result of the photovoltaic power station area includes: Obtaining a preset acquisition frequency of the fault diagnosis model; Based on the preset collection frequency, collecting the multimodal data of the photovoltaic power station area; The multimodal data is input into the fault diagnosis model to generate the fault detection result.

7. The photovoltaic power station fault diagnosis method according to claim 1, characterized in that: After the step of inputting the multimodal data into the fault diagnosis model to generate the fault detection result of the photovoltaic power station area, the method further includes: Calculating the probability of a fault occurring based on the fault detection result; Generating a warning message of a corresponding level according to the magnitude of the probability of occurrence of the fault; and Based on the fault detection result, fault area information of the photovoltaic power station area is generated.

8. The photovoltaic power station fault diagnosis method according to any one of claims 1 to 7, characterized in that: The image data includes appearance images of photovoltaic equipment in the photovoltaic power station area, and the time series data includes temperature information, humidity information, current information, voltage information of the photovoltaic equipment and meteorological information in the photovoltaic power station area.

9. A photovoltaic power station fault diagnosis device, characterized in that: The photovoltaic power station fault diagnosis device comprises: a processor and a memory, wherein instructions are stored in the memory; The processor calls the instruction in the memory so that the photovoltaic power station fault diagnosis device implements the photovoltaic power station fault diagnosis method according to any one of claims 1 to 8.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the photovoltaic power station fault diagnosis method according to any one of claims 1 to 8 is implemented.

Images