Distributed photovoltaic power prediction and control method and system

Abstract

The invention discloses a distributed photovoltaic power prediction and control method and system, and the method comprises the steps: obtaining the operation data of an inverter and the future weather forecast data of a photovoltaic power station, and carrying out the preprocessing; and inputting the preprocessed data into a neural network prediction model for prediction and outputting a power generation power prediction result, thereby realizing power prediction of different time scales of the distributed photovoltaic system. And according to a power prediction result, infrared unmanned aerial vehicle inspection, fault diagnosis early warning and photovoltaic operation and maintenance inspection work order management are combined to realize omnibearing control of the photovoltaic power station. According to the method, accurate prediction of different time scales of distributed photovoltaic power generation is realized, and safe and stable operation and high-quality power supply of a power grid are guaranteed; an intelligent photovoltaic operation and maintenance management platform is built, so that the comprehensive control of the photovoltaic power station is realized; centralized monitoring and regional maintenance are carried out on the photovoltaic power station, safe and efficient operation of the photovoltaic power station is guaranteed, operation and maintenance cost of the power station is reduced, and income of the power station is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of fault prediction and management, and in particular to a distributed photovoltaic power prediction and control method and system. Background Art

[0002] With the growth of global demand for renewable energy, photovoltaic energy has been widely valued for its clean and renewable characteristics. However, with the increase in the number of photovoltaic power stations, its operation and maintenance are facing unprecedented challenges. At present, the operation and maintenance of photovoltaic power stations mainly have the following problems: First, the degree of digitization of photovoltaic power stations is generally low. Although some large photovoltaic power stations have been equipped with data acquisition equipment and can collect a certain degree of operation data, the utilization efficiency of these data is not high, and there is a lack of effective data analysis and processing mechanisms. How to convert massive amounts of data into valuable decision-making information for fault prediction and diagnosis is a major problem currently faced. Secondly, the traditional operation and maintenance of photovoltaic power stations is highly dependent on manual operation. The operation and maintenance personnel need to conduct a comprehensive inspection of the power station regularly to ensure that each component is working properly. This operation and maintenance mode not only consumes a lot of human resources, but also has low efficiency and is prone to misjudgment or missed detection, affecting the safe and stable operation of the power station. Furthermore, the real-time monitoring and rapid response capabilities of photovoltaic power stations are weak. Due to technical and human resource limitations, many photovoltaic power stations cannot achieve all-weather real-time monitoring. Once a fault occurs, it often takes a long time from discovery to processing, which not only reduces the power generation efficiency of the power station, but also may cause economic losses.

[0003] In view of the above problems, it is urgent to develop new technical means to improve the operation and maintenance efficiency and management level of photovoltaic power stations, especially to combine advanced technologies such as the Internet of Things (IoT), Big Data (Big Data) and Artificial Intelligence (AI) to develop a set of methods and systems that can achieve accurate prediction and energy management of distributed photovoltaic power. The system should have strong data processing capabilities, be able to conduct in-depth analysis of the operating data of photovoltaic power stations, predict potential faults in advance, and provide solutions. At the same time, through intelligent monitoring means, remote monitoring and automated operation and maintenance of power stations can be achieved, reducing dependence on manual labor and improving operation and maintenance efficiency. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a distributed photovoltaic power prediction and control method and system to solve the problems that the existing photovoltaic power station operation and maintenance means are traditional and inefficient, lack effective data support and intelligent management, and are difficult to achieve timely prediction and rapid response to faults.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a distributed photovoltaic power prediction and control method, including: obtaining inverter operation data and future weather forecast data of photovoltaic power stations, and performing preprocessing; inputting the preprocessed data into a neural network prediction model for prediction and outputting the power generation prediction result, thereby realizing power prediction of distributed photovoltaics at different time scales; based on the results of the power prediction, combined with infrared drone inspections, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management, achieving all-round control of photovoltaic power stations.

[0008] As a preferred solution of the distributed photovoltaic power prediction and control method of the present invention, the acquisition of inverter operation data and future weather forecast data of the photovoltaic power station includes:

[0009] Collect the inverter operation data, including inverter AC side power, AC side voltage, AC side current, DC side power, DC side voltage, and DC side current; develop a data extraction interface to obtain real-time future weather forecast data for the location of the photovoltaic power station, including total solar irradiance, temperature, humidity, wind speed, wind direction, air pressure, and visibility;

[0010] Distributed database cluster technology is used to store the inverter operation data and future weather forecast data of the photovoltaic power station, and all real-time collected data retain the original data and back up.

[0011] As a preferred solution of the distributed photovoltaic power prediction and control method of the present invention, the preprocessing includes:

[0012] Perform integrity check and rationality check on the inverter operation data and the future weather forecast data of the photovoltaic power station, wherein the integrity check shall meet the requirements of data quantity integrity, data compliance with the start and end time ranges, and time continuity, and the rationality check shall meet the requirements of data out-of-limit check;

[0013] For missing and abnormal power data, the power data of the previous moment is used to complete it. For power data less than 0, it is uniformly replaced by 0.

[0014] Missing or abnormal meteorological data will be corrected using other meteorological elements based on linear interpolation. All corrected data will be recorded with special markings.

[0015] As a preferred solution of the distributed photovoltaic power prediction and control method of the present invention, the power prediction of distributed photovoltaic at different time scales includes:

[0016] Building a neural network prediction model based on the pytorch algorithm framework, and using the preprocessed data to train the neural network prediction model, the neural network prediction model includes a photovoltaic short-term prediction model and a photovoltaic ultra-short-term prediction model;

[0017] The input of the photovoltaic short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 24 hours, and the output is the power generation per installed capacity in the next 24 hours. The power generation prediction result for the day is calculated based on the total installed capacity, planned maintenance capacity, and planned startup capacity of the power station on the day;

[0018] The input of the photovoltaic ultra-short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 4 hours and the temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance data at the current moment, as well as the power generation data per unit installed capacity at the current moment. The output is the power data per unit installed capacity in the next 4 hours. The power generation prediction result for the next 4 hours is calculated based on the actual startup capacity counted at the current moment.

[0019] As a preferred solution of the distributed photovoltaic power prediction and control method described in the present invention, the infrared drone inspection includes:

[0020] Process thermal infrared images and visible light images, and fuse the processed images using multimodal image fusion technology;

[0021] The deep learning-based anomaly detection algorithm automatically identifies anomalies in the fused image, and uses transfer learning technology to migrate the pre-trained neural network model to the infrared drone inspection task;

[0022] According to the severity and urgency of the abnormal situation, the corresponding alarm method is automatically triggered.

[0023] As a preferred solution of the distributed photovoltaic power prediction and control method described in the present invention, the fault diagnosis and early warning includes:

[0024] According to the experience of photovoltaic power station production and operation and the requirements of photovoltaic power station construction specifications, the range of the discrete rate of the combiner box string current is classified;

[0025] If the discrete rate of the combiner box string current is within the range of 0-5%, it means that the combiner box branch current is running stably; if the discrete rate of the combiner box string current is within the range of 5%-10%, it means that the combiner box branch current is running well; if the discrete rate of the combiner box string current is within the range of 10%-20%, it means that the current operation of the combiner box branch needs to be improved; if the discrete rate of the combiner box string current is more than 20%, it means that the current operation of the combiner box branch is poor and must be rectified;

[0026] The model parameters are tuned by cross-validation to perform periodic health diagnosis on the inverter equipment.

[0027] As a preferred solution of the distributed photovoltaic power prediction and control method described in the present invention, the photovoltaic operation and maintenance inspection work order management includes:

[0028] Through back-end data entry and registration application, the organization structure of the photovoltaic power station operation and maintenance team is managed;

[0029] Carry out full life cycle management of all equipment and facilities in photovoltaic power plants, including basic equipment information entry, operation status monitoring, maintenance and repair records, and depreciation and scrap management;

[0030] Based on the above-mentioned full life cycle management, various operation and maintenance task work orders are created, dispatched, tracked and completed online.

[0031] In a second aspect, the present invention provides a distributed photovoltaic power prediction and control system, comprising:

[0032] Data acquisition and processing module, used to acquire inverter operation data and future weather forecast data of photovoltaic power station in real time and perform preprocessing;

[0033] A power prediction module is used to input the preprocessed data into a neural network prediction model for prediction and output the power generation prediction result, so as to realize power prediction of distributed photovoltaic at different time scales;

[0034] The management and control module is used to achieve all-round control of the photovoltaic power station based on the results of the power prediction, combined with infrared drone inspection, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management.

[0035] In a third aspect, the present invention provides an electronic device, comprising:

[0036] Memory and processor;

[0037] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the distributed photovoltaic power prediction and control method are implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the distributed photovoltaic power prediction and control method.

[0039] Compared with the prior art, the invention has the following beneficial effects: the invention provides a distributed photovoltaic power prediction and control method and system, which realizes accurate prediction of distributed photovoltaic power at different time scales through data collection and storage, data quality control, mesoscale NWP and power prediction functions, and provides guarantee for safe and stable operation of power grid and high-quality power supply; and through the construction of intelligent photovoltaic operation and maintenance management platform, combined with infrared drone patrol, fault self-diagnosis system, and intelligent inspection work order management function module, realizes all-round standardized, digital and refined management of photovoltaic power station. The invention conducts centralized monitoring and regional maintenance of photovoltaic power station, ensures safe and efficient operation of photovoltaic power station, and ultimately realizes the reduction of power station operation and maintenance cost and the increase of power station revenue, which is very necessary. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0041] Figure 1 A schematic diagram of the overall process logic of a distributed photovoltaic power prediction and control method according to an embodiment of the present invention;

[0042] Figure 2 A diagram of the data collection and storage architecture of a distributed photovoltaic power prediction and control method according to an embodiment of the present invention;

[0043] Figure 3 A multi-modal data recognition block diagram of a distributed photovoltaic power prediction and control method according to an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of an inverter discrete rate analysis model of a distributed photovoltaic power prediction and control method according to an embodiment of the present invention;

[0045] Figure 5 A schematic diagram of inverter health diagnosis of a distributed photovoltaic power prediction and control method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0047] Example 1

[0048] Reference Figure 1-Figure 5 As an embodiment of the present invention, a distributed photovoltaic power prediction and control method is provided, such as Figure 1 The specific steps shown include:

[0049] S100: Obtain inverter operation data and future weather forecast data of the photovoltaic power station, and perform preprocessing;

[0050] S200: input the preprocessed data into the neural network prediction model to make predictions and output the power generation prediction results, so as to realize the power prediction of distributed photovoltaic at different time scales;

[0051] S300: Based on the results of power prediction, combined with infrared drone inspections, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management, all-round control of photovoltaic power stations is achieved.

[0052] It should be noted that the present invention provides a distributed photovoltaic power prediction and control method and system, which realizes accurate prediction of distributed photovoltaic power at different time scales through data collection and storage, data quality control, mesoscale NWP and power prediction functions, and provides guarantee for safe and stable operation of power grid and high-quality power supply; and through the construction of intelligent photovoltaic operation and maintenance management platform, combined with infrared drone patrol, fault self-diagnosis system, and intelligent inspection work order management function module, realizes all-round standardized, digital and refined management of photovoltaic power station. The present invention conducts centralized monitoring and regional maintenance of photovoltaic power station, ensures safe and efficient operation of photovoltaic power station, and ultimately realizes the reduction of power station operation and maintenance costs and the increase of power station revenue, which is very necessary.

[0053] In the embodiment of the present application, the above step S100 includes the following sub-steps A1-A3;

[0054] In A1: obtaining inverter operation data and future weather forecast data of the photovoltaic power station;

[0055] In A2: perform integrity and rationality checks on the acquired inverter operation data and future weather forecast data of the photovoltaic power station;

[0056] In A3: Data management is performed on the obtained inverter operation data and future weather forecast data of the photovoltaic power station based on the inspection results.

[0057] In an optional embodiment, if Figure 2 Sub-step A1 shown specifically includes: using the Internet of Things (IoT) technology to realize real-time collection and transmission of sensor data, real-time collection of inverter operation data, including inverter AC side power, AC side voltage, AC side current, DC side power, DC side voltage, and DC side current data; based on third-party high-precision meteorological services, developing a data extraction interface to obtain real-time weather forecast data for the next 72 hours at the location of the photovoltaic power station, including total solar irradiance, temperature, humidity, wind speed, wind direction, air pressure, and visibility; the time resolution is 15 minutes. Distributed Mongo database cluster technology is used to store inverter operation data and future weather forecast data of photovoltaic power stations, and all real-time collected data retains the original data and is backed up.

[0058] In an optional embodiment, sub-step A2 specifically includes: developing a data quality assessment model based on the acquired inverter operation data and future weather forecast data of the photovoltaic power station, using isolation forest and statistical methods to perform integrity and rationality checks on the data, wherein the integrity check should meet the requirements of data quantity integrity, data compliance with the start and end time ranges, and time continuity, and the rationality check should meet the limit checks on the power generation power and meteorological data (humidity, wind speed, irradiance);

[0059] In an optional embodiment, sub-step A3 specifically includes: developing a data governance model based on the inspection results, smoothing and correcting the sensor data, and processing missing values ​​and outliers;

[0060] The processing methods and requirements include: for missing and abnormal power data, the power data of the previous moment is used to complete it; for power data less than 0, it is uniformly replaced by 0; for missing or abnormal meteorological data, other meteorological elements are used to correct it according to linear interpolation; all corrected data will be recorded with special marks; all missing and abnormal data can be supplemented or corrected manually.

[0061] It should be noted that the above step S100 not only ensures the accuracy and completeness of the data, but also improves the quality of the data by removing missing values ​​and outliers, so that the neural network prediction model can learn and predict the power generation more accurately, thereby achieving accurate prediction of distributed photovoltaics at different time scales. In addition, the preprocessing step can also integrate multi-source data, enhance the generalization ability of the model, enable it to better adapt to various complex environmental conditions, and provide reliable data support for the efficient management and safe operation of photovoltaic power stations.

[0062] In the embodiment of the present application, the above step S200 inputs the preprocessed data into the neural network prediction model for prediction and outputs the power generation prediction result, and realizes the power prediction of distributed photovoltaic at different time scales, including:

[0063] A neural network prediction model is built based on the pytorch algorithm framework, and the preprocessed data is used to train the neural network prediction model. The neural network prediction model includes a photovoltaic short-term prediction model and a photovoltaic ultra-short-term prediction model.

[0064] Specifically, the input of the photovoltaic short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 24 hours, with a time resolution of 15 minutes. The output is the power generation per installed capacity in the next 24 hours. The power generation prediction result for the day is calculated based on the total installed capacity, planned maintenance capacity, and planned startup capacity of the power station on that day.

[0065] Specifically, the input of the photovoltaic ultra-short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 4 hours and the temperature, humidity, wind speed, air pressure, visibility, total solar irradiance data at the current moment, as well as the power generation data per unit installed capacity at the current moment. The time resolution is 15 minutes, and the output is the power data per unit installed capacity in the next 4 hours. The power generation forecast result for the next 4 hours is calculated based on the actual startup capacity counted at the current moment.

[0066] It should be noted that the above step S200 can not only make full use of historical data and real-time data to improve the accuracy and reliability of prediction, but also flexibly adjust the prediction model according to the needs of different time scales, provide a scientific basis for the scheduling and operation of photovoltaic power stations, and effectively improve the power generation efficiency and economic benefits of the power station. At the same time, the learning ability of the neural network model can continuously optimize the prediction accuracy, adapt to changes in the power station operating environment, and ensure long-term stable performance.

[0067] In the embodiment of the present application, the above step S300 realizes all-round control of the photovoltaic power station based on the result of power prediction, combined with infrared drone inspection, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management, specifically including:

[0068] In an optional embodiment, the infrared drone inspection includes: processing thermal infrared images and visible light images, and fusing the processed images using multimodal image fusion technology; automatically identifying abnormalities in the fused images using a deep learning-based anomaly detection algorithm, and migrating a pre-trained neural network model to the infrared drone inspection task using transfer learning technology; and automatically triggering a corresponding alarm method according to the severity and urgency of the abnormal situation;

[0069] Specifically, the processing and analysis of infrared images includes: using infrared image preprocessing technology to reduce image noise; applying the infrared image segmentation algorithm based on region growing to segment the image into different regions; using the YOLO algorithm to automatically detect and identify abnormal components in the image;

[0070] Specifically, the processing and analysis of visible light images includes: using edge detection algorithms (such as Canny edge detection) and image processing technology to extract features in the image; using machine learning algorithms to train models for target detection and classification; and applying deep learning technology to improve the anomaly detection performance of visible light images.

[0071] Specifically, Figure 3 As shown in the figure, multimodal image fusion technology is used to fuse the processed images into a single image, and the anomaly detection algorithm based on deep learning automatically identifies anomalies in the fused image. The transfer learning technology is used to migrate the pre-trained neural network model to the infrared UAV inspection task to improve the generalization ability and detection accuracy of the model; an intelligent alarm system is designed to automatically trigger the corresponding alarm method according to the severity and urgency of the abnormal situation, such as sound, light flash, text message or mobile phone App push.

[0072] In an optional embodiment, if Figure 4 The fault diagnosis warnings shown include:

[0073] The discrete rate can be used to evaluate the consistency of the power generation performance of the power generation unit and the consistency of the string current. The discrete rate calculation formula is: discrete rate = standard deviation of string data / average value of string data * 100%; since the original data may be missing, abnormal negative values, abnormal mutations, etc., big data means are used to extract, complete, and handle abnormalities of the original data, and weight it according to the string capacity to avoid inaccurate data calculation due to capacity imbalance.

[0074] Input the processed data into the formula to calculate the discrete rate; according to the experience of photovoltaic power station production and operation and the requirements of GB50794-2012 photovoltaic power station construction specification, the range of the discrete rate of the combiner box string current can be divided into the following four levels:

[0075] ① If the discrete rate of the combiner box string current is within the range of 0-5%, it means that the combiner box branch current is stable. ② If the discrete rate of the combiner box string current is within the range of 5%-10%, it means that the combiner box branch current is operating well. ③ If the discrete rate of the combiner box string current is within the range of 10%-20%, it means that the operation of the combiner box branch current needs to be improved. ④ If the discrete rate of the combiner box branch current exceeds 20%, it means that the operation of the combiner box branch current is poor, affecting the power generation of the power station, and rectification must be carried out.

[0076] Further, such as Figure 5 As shown in the figure, big data is used to extract, complete, and handle exceptions of the inverter raw data, and to calculate the inverter conversion efficiency, average power generation time and other indicators. Combined with NWP weather data, the data and indicators are analyzed and screened, and the inverter health diagnosis model is trained based on the CNN artificial neural network algorithm. The model parameters are tuned through cross-validation and other means to improve the prediction performance. Periodic health diagnosis of the inverter equipment can be carried out to detect problems and hidden dangers of the inverter equipment in advance.

[0077] In an optional embodiment, the photovoltaic operation and maintenance inspection work order management includes: ① Online inspection work order management, first of all, the operation and maintenance engineers must be managed online, and the organizational structure of the photovoltaic power station operation and maintenance team must be managed through background entry and mobile APP registration application, including personnel information, role authority allocation, etc., so as to clearly grasp the work status and work efficiency of team members and achieve optimal allocation of human resources. ② Since operation and maintenance is actually the operation and maintenance of equipment, it is necessary to manage the entire life cycle of all equipment and facilities in the photovoltaic power station, including basic equipment information entry, operation status monitoring, maintenance and repair records, depreciation and scrap management, etc. By updating and tracking asset status in real time, data support is provided for operation and maintenance decisions. ③ It is possible to create, distribute, track and complete various types of operation and maintenance task work orders online, which may include inspection work orders, defect elimination work orders, alarm elimination work orders, and emergency repair work orders. At the same time, the system can also automatically trigger work orders according to preset rules to improve the operation and maintenance response speed and processing efficiency. Work orders can be pushed online in real time to achieve targeted and accurate distribution. Operation and maintenance personnel can receive work order notifications, view work order contents, and provide online work order submission and review through mobile APP. ④ Operation and maintenance personnel can use the operation and maintenance mobile APP to receive and push work orders, view work orders, sign in for tasks, and submit tasks. Inspection work orders can report defects, which greatly improves the efficiency of operation and maintenance management and reduces the turnover cost of operation and maintenance personnel.

[0078] It should be noted that the above step S300 can timely discover and handle equipment failures, improve the operational stability and safety of the power station, and optimize the dispatching and operation strategies of the power station through accurate power prediction, thereby improving power generation efficiency and economic benefits. In addition, the introduction of the work order management system makes the operation and maintenance work more standardized and efficient, reduces human errors, ensures the timeliness and accuracy of the operation and maintenance work, and comprehensively improves the management level of the photovoltaic power station.

[0079] Example 2

[0080] This embodiment provides a distributed photovoltaic power prediction and control system, including a data acquisition and processing module, a power prediction module and a management control module;

[0081] Specifically, the data acquisition and processing module is used to acquire the inverter operation data and the future weather forecast data of the photovoltaic power station in real time and perform preprocessing;

[0082] Specifically, the power prediction module is used to input the preprocessed data into the neural network prediction model for prediction and output the power generation prediction result, so as to realize the power prediction of distributed photovoltaic at different time scales;

[0083] Specifically, the management and control module is used to achieve all-round control of the photovoltaic power station based on the results of power prediction, combined with infrared drone inspections, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management.

[0084] It should be noted that the technical solution of the distributed photovoltaic power prediction and control system and the technical solution of the above-mentioned distributed photovoltaic power prediction and control method belong to the same concept. For the details not described in detail in the technical solution of the distributed photovoltaic power prediction and control system in this embodiment, please refer to the description of the technical solution of the above-mentioned distributed photovoltaic power prediction and control method.

[0085] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0086] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a distributed photovoltaic power prediction and control method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0087] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method proposed in the above embodiment is implemented.

[0088] The storage medium proposed in this embodiment and the method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0089] Through the above description of the implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation mode. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the method of the embodiment of the present invention.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0091] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0092] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0095] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0096] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A distributed photovoltaic power prediction and control method, characterized in that: include: Obtain inverter operation data and future weather forecast data of photovoltaic power stations and perform preprocessing; The pre-processed data is input into a neural network prediction model to perform prediction and output a power generation prediction result, thereby realizing power prediction of distributed photovoltaics at different time scales; According to the results of the power prediction, combined with infrared drone inspection, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management, all-round control of the photovoltaic power station is achieved.

2. The distributed photovoltaic power prediction and control method according to claim 1, characterized in that: The acquisition of inverter operation data and future weather forecast data of the photovoltaic power station includes: Collect the inverter operation data, including inverter AC side power, AC side voltage, AC side current, DC side power, DC side voltage, and DC side current; develop a data extraction interface to obtain real-time future weather forecast data for the location of the photovoltaic power station, including total solar irradiance, temperature, humidity, wind speed, wind direction, air pressure, and visibility; Distributed database cluster technology is used to store the inverter operation data and future weather forecast data of the photovoltaic power station, and all real-time collected data retain the original data and back up.

3. The distributed photovoltaic power prediction and control method according to claim 2, characterized in that: The pre-processing comprises: Perform integrity check and rationality check on the inverter operation data and the future weather forecast data of the photovoltaic power station, wherein the integrity check shall meet the requirements of data quantity integrity, data compliance with the start and end time ranges, and time continuity, and the rationality check shall meet the requirements of data out-of-limit check; For missing and abnormal power data, the power data of the previous moment is used to complete it. For power data less than 0, it is uniformly replaced by 0. Missing or abnormal meteorological data will be corrected using other meteorological elements based on linear interpolation. All corrected data will be recorded with special markings.

4. The distributed photovoltaic power prediction and control method according to claim 3, characterized in that: The power prediction of distributed photovoltaic at different time scales includes: Building a neural network prediction model based on the pytorch algorithm framework, and using the preprocessed data to train the neural network prediction model, the neural network prediction model includes a photovoltaic short-term prediction model and a photovoltaic ultra-short-term prediction model; The input of the photovoltaic short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 24 hours, and the output is the power generation per installed capacity in the next 24 hours. The power generation prediction result for the day is calculated based on the total installed capacity, planned maintenance capacity, and planned startup capacity of the power station on the day; The input of the photovoltaic ultra-short-term prediction model is the characteristic data of temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance in the next 4 hours and the temperature, humidity, wind speed, air pressure, visibility, and total solar irradiance data at the current moment, as well as the power generation data per unit installed capacity at the current moment. The output is the power data per unit installed capacity in the next 4 hours. The power generation prediction result for the next 4 hours is calculated based on the actual startup capacity counted at the current moment.

5. The distributed photovoltaic power prediction and control method according to claim 4, characterized in that: The infrared drone inspection includes: Process thermal infrared images and visible light images, and fuse the processed images using multimodal image fusion technology; The deep learning-based anomaly detection algorithm automatically identifies anomalies in the fused image, and uses transfer learning technology to migrate the pre-trained neural network model to the infrared drone inspection task; According to the severity and urgency of the abnormal situation, the corresponding alarm method is automatically triggered.

6. The distributed photovoltaic power prediction and control method according to claim 5, characterized in that: The fault diagnosis and early warning comprises: According to the experience of photovoltaic power station production and operation and the requirements of photovoltaic power station construction specifications, the range of the discrete rate of the combiner box string current is classified; If the discrete rate of the combiner box string current is within the range of 0-5%, it means that the combiner box branch current is running stably; if the discrete rate of the combiner box string current is within the range of 5%-10%, it means that the combiner box branch current is running well; if the discrete rate of the combiner box string current is within the range of 10%-20%, it means that the current operation of the combiner box branch needs to be improved; if the discrete rate of the combiner box string current is more than 20%, it means that the current operation of the combiner box branch is poor and must be rectified; The model parameters are tuned by cross-validation to perform periodic health diagnosis on the inverter equipment.

7. The distributed photovoltaic power prediction and control method according to claim 6, characterized in that: The photovoltaic operation and maintenance inspection work order management includes: Through back-end data entry and registration application, the organization structure of the photovoltaic power station operation and maintenance team is managed; Carry out full life cycle management of all equipment and facilities in photovoltaic power plants, including basic equipment information entry, operation status monitoring, maintenance and repair records, and depreciation and scrap management; Based on the above-mentioned full life cycle management, various operation and maintenance task work orders are created, dispatched, tracked and completed online.

8. A system using the distributed photovoltaic power prediction and control method according to any one of claims 1 to 7, characterized in that: include: Data acquisition and processing module, used to acquire inverter operation data and future weather forecast data of photovoltaic power station in real time and perform preprocessing; A power prediction module is used to input the preprocessed data into a neural network prediction model for prediction and output the power generation prediction result, so as to realize power prediction of distributed photovoltaic at different time scales; The management and control module is used to achieve all-round control of the photovoltaic power station based on the results of the power prediction, combined with infrared drone inspection, fault diagnosis and early warning, and photovoltaic operation and maintenance inspection work order management.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the distributed photovoltaic power prediction and control method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the distributed photovoltaic power prediction and control method according to any one of claims 1 to 7.

Images