Intelligent photovoltaic power station operation and maintenance method, system, equipment and medium

Abstract

The invention provides an intelligent photovoltaic power station operation and maintenance method, system and device and a medium, and belongs to the technical field of photovoltaic power station operation and maintenance, and the method comprises the following steps: collecting historical operation data and meteorological data of a photovoltaic power station, building a database, and marking the operation state; establishing a fault identification model and a generating capacity prediction model for the photovoltaic power station, and establishing a data set according to the operation data of the photovoltaic power station; using the data set to train a fault identification model and a generating capacity prediction model, using the fault identification model to identify the fault and the fault type of each photovoltaic power station, and using the generating capacity prediction model to predict the generating capacity of each photovoltaic power station; and the power grid side demand electric quantity is obtained, the redundant generating capacity is calculated according to the predicted generating capacity of the photovoltaic power station and the faulted photovoltaic power station, and the corresponding photovoltaic power station is selected for maintenance according to the redundant generating capacity and the maintenance record. Intelligent operation and maintenance of the photovoltaic power station are realized, the operation and maintenance efficiency and accuracy are improved, and the operation and maintenance cost is reduced.

Description

Background Art

[0002] Photovoltaic power generation is a new energy power generation system that directly converts solar radiation energy into electrical energy based on the photovoltaic effect of semiconductor materials. As a type of renewable energy, photovoltaic power generation has become a powerful supplement to traditional power generation. It has been promoted due to its renewable sustainability, environmental protection, flexible energy distribution, safety and reliability. With the continuous maturity of photovoltaic technology, the scale of photovoltaic power station construction has continued to expand, and the level of photovoltaic power generation has been significantly improved. The operation and maintenance of photovoltaic power stations is an important part of the operation process of photovoltaic power stations.

[0003] The traditional operation and maintenance of photovoltaic power stations is done by manual inspection. As the scale of photovoltaic power stations grows, the disadvantages of this method are becoming more and more apparent, mainly in the following aspects: First, the operation and maintenance cost is high. Power stations are widely distributed and have a large number of equipment. A large amount of inspection, cleaning and maintenance work requires a lot of manpower and material resources, and requires high technical capabilities of operation and maintenance personnel. Second, the operation and maintenance efficiency is low. The manual inspection method is not only inefficient, but also prone to missed inspections and false inspections. With the development of Internet of Things technology, a new generation of technology for remote monitoring of photovoltaic power stations through sensors has emerged, but the lack of effective data supervision has resulted in the inability of operation and maintenance personnel to accurately understand the operating status of photovoltaic power stations, let alone potential operating risks.

[0004] This is a shortcoming of the prior art. Therefore, it is very necessary to provide an intelligent photovoltaic power station operation and maintenance method, system, equipment and medium to address the above-mentioned defects in the prior art. Summary of the invention

[0005] In view of the high cost and low efficiency of traditional widespread power station operation and maintenance, and the lack of effective data monitoring for remote monitoring of photovoltaic power stations through sensors, which makes it impossible to accurately or widely monitor the operating conditions of power stations, the present invention provides an intelligent photovoltaic power station operation and maintenance method, system, equipment and medium to solve the above technical problems.

[0006] In a first aspect, the present invention provides an intelligent photovoltaic power station operation and maintenance method, comprising the following steps: S1. Collect historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; S2. Establishing a fault identification model and a power generation prediction model for a photovoltaic power station, establishing a first data set based on the photovoltaic power station operation data and the marked operation status, and establishing a second data set based on the photovoltaic power station operation data; S3. Using the first data set to train the fault identification model, and using the second data set to train the power generation prediction model, and then using the trained fault identification model to identify the faults of each photovoltaic power station and the corresponding fault type, and using the trained power generation prediction model to predict the power generation of each photovoltaic power station; S4. Obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records. Furthermore, the specific steps of step S1 are as follows: S11. Collect historical operation data of each photovoltaic power station and record historical meteorological data corresponding to the historical operation data; the historical operation data includes photovoltaic module temperature, photovoltaic module current value, photovoltaic module voltage value, photovoltaic power station power generation, photovoltaic module surface dust, the historical meteorological data includes light intensity, light time, temperature; S12. Establish a database to store historical operation data and historical meteorological data of each photovoltaic power station; S13. Annotate the operation status of the photovoltaic power station corresponding to the historical operation data in the database, where the operation status includes normal and fault.

[0007] Furthermore, the specific steps of step S2 are as follows: S21. Obtain historical operation data of the photovoltaic power station and classify it into normal and fault states; S22. Further label the fault type label on the historical operation data of the fault state; S23. Establish a fault identification model for the photovoltaic power station based on a convolutional neural network, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust and photovoltaic module temperature in the historical operation data as inputs of the fault identification model, and use the fault type as output of the photovoltaic identification model; S24. Establishing a first data set based on the historical operation data of the photovoltaic power station and the operation status and fault type of the photovoltaic power station; S25. Establish a power generation prediction model for the photovoltaic power station based on the BP neural network, determine the number of network layers and nodes, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust, photovoltaic module temperature in the historical operation data, and the light intensity and light time in the historical meteorological data as inputs of the power generation prediction model, and use the power generation of the photovoltaic power station in the historical operation data as the output of the power generation prediction model; S26. Establish a second data set based on the historical operation data of the photovoltaic power station and the corresponding historical meteorological data.

[0008] Furthermore, the specific steps of step S3 are as follows: S31. Filter out the first training data and the first test data from the first data set according to the set ratio, initialize the network parameters of the fault identification model, set the loss function, the maximum number of iterations, and the convergence threshold; S32. Preprocessing the first training data and inputting it into the network of the fault recognition model for training, and optimizing the network parameters by the back propagation algorithm during the training process; S33. Evaluate the trained fault recognition model using the first test data, adjust the fault recognition model according to the evaluation result, and obtain a fault recognition model that passes the evaluation; S34. Filter out the second training data and the second test data from the second data set according to the set ratio, and initialize the network parameters, activation function, loss function, maximum number of iterations, convergence threshold, and learning rate of the power generation prediction model; S35. Preprocess the second training data and input it into the power generation prediction model for training until the number of iterations is greater than the maximum number of iterations or the model converges to obtain the optimal power generation prediction model; S36. Testing the optimal power generation prediction model using the second test data set, and returning to step S35 to retrain if the test fails, and obtaining the final power generation prediction model if the test passes; S37. Collect real-time photovoltaic power station operation data, input it into the fault identification model that has passed the evaluation, determine whether the operation status of the photovoltaic power station is normal, and output the fault type in the event of a fault; S38. Collect real-time photovoltaic power station operation data and meteorological data, input them into the final power generation prediction model, and determine the power generation prediction value of each photovoltaic power station.

[0009] Furthermore, the specific steps of step S4 are as follows: S41. Obtain the power demand W on the grid side X ; S42. Obtain the predicted power generation value of each photovoltaic power station and sum them up to obtain the predicted total power generation W TP ; S43. Determine the failed photovoltaic power station and determine whether the failed photovoltaic power station is still in operation; If yes, go to step S44; If not, the maximum power generation capacity W K =W TP , proceed to step S45; S44. Obtain the predicted power generation value W of the faulty photovoltaic power station GP , then the maximum power generation W K =W TP -W GP ; S45. Calculate the redundant power generation WR =W K -W X ; S46. Obtain maintenance records of each photovoltaic power station and establish a maintenance record database; S47. Selecting the photovoltaic power station with the longest time from the last maintenance to the current time from the maintenance record database as the pre-maintenance photovoltaic power station; S48. Determine the predicted power generation value W of the pre-maintenance photovoltaic power station BY , judge W BY Is it less than or equal to W? R ; If yes, shut down the pre-maintenance photovoltaic power station and perform maintenance, and return to step S37; If not, the pre-maintenance photovoltaic power station is deleted from the maintenance record database and the process returns to step S46.

[0010] Furthermore, the specific steps of step S45 are as follows: S451. Compare the maximum power generation W K The power demand W on the grid side X ; When W K >W X , proceed to step S452; When W K =W X , proceed to step S453; When W K <W X , proceed to step S454; S452. Calculate the redundant power generation W R =W K -W X , proceed to step S46; S453. Determine whether the current power generation of the photovoltaic power station can meet the grid-side demand, and return to step S37; S454. Calculate the power shortage W Q = W X -W K , start the energy storage system to supplement the power supply to the grid side, and return to step S37.

[0011] Furthermore, the specific steps of step S47 are as follows: S471. Obtain the last maintenance time of each photovoltaic power station in the maintenance record database; S472. Determine whether the difference between the last maintenance time and the current time of each photovoltaic power station is less than the minimum maintenance time threshold; If yes, go to step S473; If not, proceed to step S474; S473. Use redundant power generation W R Charge the energy storage system and return to step S37; S474. Delete the photovoltaic power stations whose difference between the last maintenance time and the current time is less than or equal to the minimum maintenance time threshold from the maintenance record database, and select the photovoltaic power station whose last maintenance time is the longest from the current time as the pre-maintenance photovoltaic power station.

[0012] In a second aspect, the present invention provides an intelligent photovoltaic power station operation and maintenance system, comprising: The data acquisition module is used to collect the historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; A model building module, used to establish a fault identification model and a power generation prediction model for the photovoltaic power station, establish a first data set according to the photovoltaic power station operation data and the marked operation status, and establish a second data set according to the photovoltaic power station operation data; A model training module, used to train a fault identification model using the first data set, and to train a power generation prediction model using the second data set, and then to use the trained fault identification model to identify faults and corresponding fault types of each photovoltaic power station, and to use the trained power generation prediction model to predict the power generation of each photovoltaic power station; The intelligent maintenance module of the photovoltaic power station is used to obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records.

[0013] In a third aspect, the present invention provides an electronic device, including a processor and a memory; The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the electronic device executes the method described in the first aspect above.

[0014] In a fourth aspect, the present invention provides a storage medium, The storage medium stores instructions, which, when executed on a computer, enable the computer to execute the method described in the first aspect.

[0015] The beneficial effects of the present invention are: The intelligent photovoltaic power station operation and maintenance method, system, equipment and medium provided by the present invention collect historical operation data and meteorological data of the photovoltaic power station, establish a database and mark the operation status, and then establish a fault identification model and a power generation prediction model, thereby realizing intelligent operation and maintenance of the photovoltaic power station, improving operation and maintenance efficiency and accuracy, and reducing operation and maintenance costs.

[0016] It can be seen that compared with the prior art, the present invention has outstanding substantive features and significant progress, and the beneficial effects of its implementation are also obvious. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 It is a flow chart of the intelligent photovoltaic power station operation and maintenance method of the present invention.

[0019] Figure 2 It is a schematic diagram of the intelligent photovoltaic power station operation and maintenance system of the present invention. DETAILED DESCRIPTION

[0020] The intelligent photovoltaic power station operation and maintenance method can be applied to one or more electronic devices, wherein the electronic device is an electronic device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0021] An electronic device may be any electronic product that can interact with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), an interactive network television (IPTV), etc. The network in which the electronic device is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0022] 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 should fall within the scope of protection of the present invention.

[0023] See also Figure 1 The figure shows a specific implementation of an intelligent photovoltaic power station operation and maintenance method, including the following steps: S1. Collect historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; S2. Establishing a fault identification model and a power generation prediction model for a photovoltaic power station, establishing a first data set based on the photovoltaic power station operation data and the marked operation status, and establishing a second data set based on the photovoltaic power station operation data; S3. Using the first data set to train the fault identification model, and using the second data set to train the power generation prediction model, and then using the trained fault identification model to identify the faults of each photovoltaic power station and the corresponding fault type, and using the trained power generation prediction model to predict the power generation of each photovoltaic power station; S4. Obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records.

[0024] In this embodiment, the specific steps of step S1 are as follows: S11. Collect historical operation data of each photovoltaic power station and record historical meteorological data corresponding to the historical operation data; the historical operation data includes photovoltaic module temperature, photovoltaic module current value, photovoltaic module voltage value, photovoltaic power station power generation, photovoltaic module surface dust, the historical meteorological data includes light intensity, light time, temperature; S12. Establish a database to store historical operation data and historical meteorological data of each photovoltaic power station; S13. Annotate the operating status of the photovoltaic power station corresponding to the historical operating data in the database, where the operating status includes normal and fault; It should be noted that this provides accurate and comprehensive data support for the establishment of subsequent models; The specific steps of step S2 are as follows: S21. Obtain historical operation data of the photovoltaic power station and classify it into normal and fault states; S22. Further label the fault type labels for the historical operation data of the fault state; the fault type labels include hot spot fault, aging fault, potential induced degradation fault, and shielding fault; S23. Establish a fault identification model for the photovoltaic power station based on a convolutional neural network, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust and photovoltaic module temperature in the historical operation data as inputs of the fault identification model, and use the fault type as output of the photovoltaic identification model; Specifically, the characteristic graph of the photovoltaic module current value, the photovoltaic module voltage value, the surface dust of the photovoltaic module and the temperature of the photovoltaic module in the historical operation data is used as input, and the IV curve of the photovoltaic module is the characteristic graph corresponding to the photovoltaic module current value and the photovoltaic module voltage value; The following fault identification model is established: C = g ( X ′, V , c ) Wherein, C is the predicted fault type, which can be hot spot fault, aging fault, potential induced degradation fault, and shielding fault; X ′ is the preprocessed feature vector, which is the feature map of the PV module current value, PV module voltage value, surface dust of PV module and temperature of PV module; V is the weight matrix of the convolution layer; c is the bias vector of the convolutional layer, g It is a composite function of convolutional neural network, including convolutional layer, pooling layer and fully connected layer; S24. Establishing a first data set based on the historical operation data of the photovoltaic power station and the operation status and fault type of the photovoltaic power station; S25. Establish a power generation prediction model for the photovoltaic power station based on the BP neural network, determine the number of network layers and nodes, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust, photovoltaic module temperature in the historical operation data, and the light intensity and light time in the historical meteorological data as inputs of the power generation prediction model, and use the power generation of the photovoltaic power station in the historical operation data as the output of the power generation prediction model; Specifically, the following power generation prediction model is established: Y = f ( X , W , b ) Where Y is the predicted power generation of the PV plant; X is the input feature vector, including the PV module current value, PV module voltage value, PV module surface dust, PV module temperature, light intensity and light time; W is the weight matrix of the BP neural network, b is the bias vector of the BP neural network, f It is the activation function of the BP neural network, such as the Sigmoid nonlinear function; S26. Establishing a second data set based on the historical operation data of the photovoltaic power station and the corresponding historical meteorological data; It should be noted that the accuracy and reliability of the model are ensured here, which is the basis for subsequent training and application; The specific steps of step S3 are as follows: S31. Filter out the first training data and the first test data from the first data set according to the set ratio, initialize the network parameters of the fault identification model, set the loss function, the maximum number of iterations, and the convergence threshold; S32. Preprocessing the first training data and inputting it into the network of the fault recognition model for training, and optimizing the network parameters by the back propagation algorithm during the training process; S33. Evaluate the trained fault recognition model using the first test data, adjust the fault recognition model according to the evaluation result, and obtain a fault recognition model that passes the evaluation; Specifically, the convolutional layer weight matrix is ​​randomly initialized V and the bias vector c , the output of the convolutional layer a conv It can be expressed as: a conv = f (Conv2D( X ′, V )+ c ) Among them, Conv2D represents a two-dimensional convolution operation, V is the weight matrix of the convolution kernel, c is the bias vector of the convolutional layer; Output of the pooling layer a pool It can be expressed as: a pool =Pooling( a conv ) Output of the fully connected layer a fc It can be expressed as: z fc = W fc a pool + b fc a fc = f ( z fc ) in, W fc is the weight matrix of the fully connected layer, b fc is the bias vector of the fully connected layer; For the output layer, the predicted fault type C can be expressed as: C =softmax( z out ) in z out is the weighted sum of the output layer; Set the cross entropy loss function to predict the fault type C and the actual fault type C true The error between:

[0025] in K is the number of fault types; The back propagation algorithm is used to calculate the gradient of the error with respect to the weights of each layer, and the weights and biases are updated; Iterate the training until the performance of the fault recognition model on the first test data is no longer improved or the preset number of iterations is reached to obtain the final fault recognition model; S34. Filter out the second training data and the second test data from the second data set according to the set ratio, and initialize the network parameters, activation function, loss function, maximum number of iterations, convergence threshold, and learning rate of the power generation prediction model; S35. Preprocess the second training data and input it into the power generation prediction model for training until the number of iterations is greater than the maximum number of iterations or the model converges to obtain the optimal power generation prediction model; S36. Testing the optimal power generation prediction model using the second test data set, and returning to step S35 to retrain if the test fails, and obtaining the final power generation prediction model if the test passes; Specifically, the network's weight matrix W and bias vector b are initialized by generating random numbers that follow a normal distribution; For input data X , the output of each layer is calculated by weighted sum and activation function, assuming that the network has L Layer, l Output of the layer al It can be expressed as: z l = W l a l−1 + b l a l = f (z l ) in, z l It is l The weighted sum of the layers, W l It is l Layer to l +1 layer weight matrix, b l It is l The bias vector of the layer, f is the activation function; For the output layer, the predicted value Y It can be expressed as: Y = a L Set the mean square error as the loss function to calculate the predicted value Y and the true value Y true The error between:

[0026] Where N is the number of samples; Use the chain rule to calculate the gradient of the error with respect to the weights of each layer:

[0027] delta l−1 =( W l ) T δ l f ′( zl−1 ) Then update the weights and biases: W l =W l −ηδ l (a l−1 ) T

[0028] in, is the learning rate, m is the number of neurons in each layer; Iterate the training until the maximum number of iterations is reached or the error is less than a preset threshold, and then verify the final power generation prediction model through the second test data; S37. Collect real-time photovoltaic power station operation data, input it into the fault identification model that has passed the evaluation, determine whether the operation status of the photovoltaic power station is normal, and output the fault type in the event of a fault; S38. Collect real-time photovoltaic power station operation data and meteorological data, input them into the final power generation prediction model, and determine the power generation prediction value of each photovoltaic power station; It should be noted that the accuracy and stability of the fault identification model and power generation prediction model were ensured through rigorous training and testing; The specific steps of step S4 are as follows: S41. Obtain the power demand W on the grid side X ; S42. Obtain the predicted power generation value of each photovoltaic power station and sum them up to obtain the predicted total power generation W TP ; S43. Determine the failed photovoltaic power station and determine whether the failed photovoltaic power station is still in operation; If yes, go to step S44; If not, the maximum power generation capacity W K =W TP , proceed to step S45; S44. Obtain the predicted power generation value W of the faulty photovoltaic power station GP , then the maximum power generation W K =W TP -W GP ; S45. Calculate the redundant power generation W R =W K -W X ; S46. Obtain maintenance records of each photovoltaic power station and establish a maintenance record database; S47. Selecting the photovoltaic power station with the longest time from the last maintenance to the current time from the maintenance record database as the pre-maintenance photovoltaic power station; S48. Determine the predicted power generation value W of the pre-maintenance photovoltaic power station BY , judge W BY Is it less than or equal to W? R ; If yes, shut down the pre-maintenance photovoltaic power station and perform maintenance, and return to step S37; It should be noted that W also needs to be calculated here R With W BY If ΔW is still greater than the predicted power generation value W of a pre-maintenance PV power station, the difference ΔW is returned to the maintenance record database. BY , the PV power station can continue to be shut down and maintained until ΔW cannot meet the power generation forecast value W of any pre-maintenance PV power station BY ; If not, the pre-maintenance photovoltaic power station is deleted from the maintenance record database, and the process returns to step S46; It should be noted that the process of calculating the redundant power generation based on the predicted power generation of the photovoltaic power station and the power demand on the grid side, and selecting the corresponding photovoltaic power station for maintenance based on the maintenance records, realizes the intelligent maintenance of the photovoltaic power station and improves the targetedness and efficiency of maintenance.

[0029] In some embodiments, step S45 includes the following steps: S451. Compare the maximum power generation W K The power demand W on the grid side X ; When W K >W X , proceed to step S452; When W K =W X , proceed to step S453; When W K <W X , proceed to step S454; S452. Calculate the redundant power generation W R =W K -W X , proceed to step S46; S453. Determine whether the current power generation of the photovoltaic power station can meet the grid-side demand, and return to step S37; S454. Calculate the power shortage W Q = W X -W K , start the energy storage system as a supplement to supply power to the grid side, and return to step S37; It should be noted that the calculation process of the redundant power generation is given here. By comparing with the power demand on the grid side and starting the energy storage system, the stable power supply of the grid and the efficient operation of the photovoltaic power station are ensured; The specific steps of step S47 are as follows: S471. Obtain the last maintenance time of each photovoltaic power station in the maintenance record database; S472. Determine whether the difference between the last maintenance time and the current time of each photovoltaic power station is less than the minimum maintenance time threshold; If yes, go to step S473; If not, proceed to step S474; S473. Use redundant power generation W R Charge the energy storage system and return to step S37; S474. Delete the photovoltaic power station whose difference between the last maintenance time and the current time is less than or equal to the minimum maintenance time threshold from the maintenance record database, and select the photovoltaic power station whose last maintenance time is the longest from the current time as the pre-maintenance photovoltaic power station; It should be noted that the process of selecting the pre-maintenance photovoltaic power station is limited here, including the judgment of the maintenance time and the charging of the energy storage system, which ensures the timely maintenance and long-term stable operation of the photovoltaic power station.

[0030] It should be understood that the order of execution of the steps in the above embodiment does not mean that the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0031] The following is an embodiment of the intelligent photovoltaic power station operation and maintenance system provided by the embodiments of the present disclosure. The system and the intelligent photovoltaic power station operation and maintenance methods of the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the intelligent photovoltaic power station operation and maintenance system, reference can be made to the embodiments of the above-mentioned intelligent photovoltaic power station operation and maintenance methods.

[0032] like Figure 2 As shown, the system includes: The data acquisition module is used to collect the historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; A model building module, used to establish a fault identification model and a power generation prediction model for the photovoltaic power station, establish a first data set according to the photovoltaic power station operation data and the marked operation status, and establish a second data set according to the photovoltaic power station operation data; A model training module, used to train a fault identification model using the first data set, and to train a power generation prediction model using the second data set, and then to use the trained fault identification model to identify faults and corresponding fault types of each photovoltaic power station, and to use the trained power generation prediction model to predict the power generation of each photovoltaic power station; The intelligent maintenance module of the photovoltaic power station is used to obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records.

[0033] The intelligent photovoltaic power station operation and maintenance method provided by the present invention is a combination of the modules and algorithm steps of each example described in the embodiments disclosed herein, and can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0034] Those skilled in the art will appreciate that the various aspects of the intelligent photovoltaic power station operation and maintenance method provided by the present invention can be implemented as a system, method or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software, which can be collectively referred to as "circuit", "module" or "system" here.

[0035] In the non-transitory computer-readable storage medium storing the intelligent photovoltaic power station operation and maintenance method, it can be but not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0036] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. A smart photovoltaic power station operation and maintenance method, characterized in that: The steps include: S1. Collect historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; S2. Establishing a fault identification model and a power generation prediction model for a photovoltaic power station, establishing a first data set based on the photovoltaic power station operation data and the marked operation status, and establishing a second data set based on the photovoltaic power station operation data; S3. Using the first data set to train the fault identification model, and using the second data set to train the power generation prediction model, and then using the trained fault identification model to identify the faults of each photovoltaic power station and the corresponding fault type, and using the trained power generation prediction model to predict the power generation of each photovoltaic power station; S4. Obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records.

2. The intelligent photovoltaic power station operation and maintenance method according to claim 1, characterized in that: The specific steps of step S1 are as follows: S11. Collect historical operation data of each photovoltaic power station and record historical meteorological data corresponding to the historical operation data; the historical operation data includes photovoltaic module temperature, photovoltaic module current value, photovoltaic module voltage value, photovoltaic power station power generation, photovoltaic module surface dust, the historical meteorological data includes light intensity, light time, temperature; S12. Establish a database to store historical operation data and historical meteorological data of each photovoltaic power station; S13. Annotate the operation status of the photovoltaic power station corresponding to the historical operation data in the database, where the operation status includes normal and fault.

3. The intelligent photovoltaic power station operation and maintenance method according to claim 2, characterized in that: The specific steps of step S2 are as follows: S21. Obtain historical operation data of the photovoltaic power station and classify it into normal and fault states; S22. Further label the fault type label on the historical operation data of the fault state; S23. Establish a fault identification model for the photovoltaic power station based on a convolutional neural network, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust and photovoltaic module temperature in the historical operation data as inputs of the fault identification model, and use the fault type as output of the photovoltaic identification model; S24. Establishing a first data set based on the historical operation data of the photovoltaic power station and the operation status and fault type of the photovoltaic power station; S25. Establish a power generation prediction model for the photovoltaic power station based on the BP neural network, determine the number of network layers and nodes, use the photovoltaic module current value, photovoltaic module voltage value, photovoltaic module surface dust, photovoltaic module temperature in the historical operation data, and the light intensity and light time in the historical meteorological data as inputs of the power generation prediction model, and use the power generation of the photovoltaic power station in the historical operation data as the output of the power generation prediction model; S26. Establish a second data set based on the historical operation data of the photovoltaic power station and the corresponding historical meteorological data.

4. The intelligent photovoltaic power station operation and maintenance method according to claim 3, characterized in that: The specific steps of step S3 are as follows: S31. Filter out the first training data and the first test data from the first data set according to the set ratio, initialize the network parameters of the fault identification model, set the loss function, the maximum number of iterations, and the convergence threshold; S32. Preprocessing the first training data and inputting it into the network of the fault recognition model for training, and optimizing the network parameters by the back propagation algorithm during the training process; S33. Evaluate the trained fault recognition model using the first test data, adjust the fault recognition model according to the evaluation result, and obtain a fault recognition model that passes the evaluation; S34. Filter out the second training data and the second test data from the second data set according to the set ratio, and initialize the network parameters, activation function, loss function, maximum number of iterations, convergence threshold, and learning rate of the power generation prediction model; S35. Preprocess the second training data and input it into the power generation prediction model for training until the number of iterations is greater than the maximum number of iterations or the model converges to obtain the optimal power generation prediction model; S36. Testing the optimal power generation prediction model using the second test data set, and returning to step S35 to retrain if the test fails, and obtaining the final power generation prediction model if the test passes; S37. Collect real-time photovoltaic power station operation data, input it into the fault identification model that has passed the evaluation, determine whether the operation status of the photovoltaic power station is normal, and output the fault type in the event of a fault; S38. Collect real-time photovoltaic power station operation data and meteorological data, input them into the final power generation prediction model, and determine the power generation prediction value of each photovoltaic power station.

5. The intelligent photovoltaic power station operation and maintenance method according to claim 4, characterized in that: The specific steps of step S4 are as follows: S41. Obtain the power demand W on the grid side X ; S42. Obtain the predicted power generation value of each photovoltaic power station and sum them up to obtain the predicted total power generation W TP ; S43. Determine the failed photovoltaic power station and determine whether the failed photovoltaic power station is still in operation; If yes, go to step S44; If not, the maximum power generation capacity W K =W TP , proceed to step S45; S44. Obtain the predicted power generation value W of the faulty photovoltaic power station GP , then the maximum power generation W K =W TP -W GP ; S45. Calculate the redundant power generation W R =W K -W X ; S46. Obtain maintenance records of each photovoltaic power station and establish a maintenance record database; S47. Selecting the photovoltaic power station with the longest time from the last maintenance to the current time from the maintenance record database as the pre-maintenance photovoltaic power station; S48. Determine the predicted power generation value W of the pre-maintenance photovoltaic power station BY , judge W BY Is it less than or equal to W? R ; If yes, shut down the pre-maintenance photovoltaic power station and perform maintenance, and return to step S37; If not, the pre-maintenance photovoltaic power station is deleted from the maintenance record database and the process returns to step S46.

6. The intelligent photovoltaic power station operation and maintenance method according to claim 5, characterized in that: The specific steps of step S45 are as follows: S451. Compare the maximum power generation W K The power demand W on the grid side X ; When W K >W X , proceed to step S452; When W K =W X , proceed to step S453; When W K <W X , proceed to step S454; S452. Calculate the redundant power generation W R =W K -W X , proceed to step S46; S453. Determine whether the current power generation of the photovoltaic power station can meet the grid-side demand, and return to step S37; S454. Calculate the power shortage W Q = W X -W K , start the energy storage system to supplement the power supply to the grid side, and return to step S37.

7. The intelligent photovoltaic power station operation and maintenance method according to claim 6, characterized in that: The specific steps of step S47 are as follows: S471. Obtain the last maintenance time of each photovoltaic power station in the maintenance record database; S472. Determine whether the difference between the last maintenance time and the current time of each photovoltaic power station is less than the minimum maintenance time threshold; If yes, go to step S473; If not, proceed to step S474; S473. Use redundant power generation W R Charge the energy storage system and return to step S37; S474. Delete the photovoltaic power stations whose difference between the last maintenance time and the current time is less than or equal to the minimum maintenance time threshold from the maintenance record database, and select the photovoltaic power station whose last maintenance time is the longest from the current time as the pre-maintenance photovoltaic power station.

8. An intelligent photovoltaic power station operation and maintenance system, characterized in that: include: The data acquisition module is used to collect the historical operation data and historical meteorological data of the photovoltaic power station to establish a database, and mark the operation status of the photovoltaic power station in the database; A model building module, used to establish a fault identification model and a power generation prediction model for the photovoltaic power station, establish a first data set according to the photovoltaic power station operation data and the marked operation status, and establish a second data set according to the photovoltaic power station operation data; A model training module, used to train a fault identification model using the first data set, and to train a power generation prediction model using the second data set, and then to use the trained fault identification model to identify faults and corresponding fault types of each photovoltaic power station, and to use the trained power generation prediction model to predict the power generation of each photovoltaic power station; The intelligent maintenance module of the photovoltaic power station is used to obtain the power demand on the grid side, calculate the redundant power generation according to the predicted power generation of the photovoltaic power station and the faulty photovoltaic power station, and select the corresponding photovoltaic power station for maintenance according to the redundant power generation and maintenance records.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the electronic device executes the method described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores instructions, which, when executed on a computer, enable the computer to execute the method described in any one of claims 1 to 7.

Images