Distributed photovoltaic power station intelligent management method and system based on big data monitoring

Abstract

The invention discloses a distributed photovoltaic power station intelligent management method and system based on big data monitoring, and the method comprises the steps: constructing a distributed photovoltaic power station big data platform, and enabling the big data platform to receive the basic information, historical operation state data and historical operation and maintenance data of various types of photovoltaic equipment, an operation state definition model and a fault type evaluation model of each type of photovoltaic equipment are constructed, an operation and maintenance plan of a first operation and maintenance period is formulated and executed based on basic information and historical operation and maintenance data of each type of photovoltaic equipment, and operation and maintenance data of the first operation and maintenance period are uploaded to a big data platform during the period to be updated and adjusted timely. The adjusted operation and maintenance plan becomes a guidance basis of a second operation and maintenance period, continuous optimization of operation and maintenance management is realized, an operation state evaluation model and a fault type evaluation model of the photovoltaic equipment are updated in real time based on data in a big data platform and are closer to the actual operation condition of the equipment, the adaptability and accuracy of the model are improved, and the operation and maintenance efficiency is improved. And a better use prospect is brought.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to an intelligent management method and system for distributed photovoltaic power stations based on big data monitoring. Background Art

[0002] At present, the global energy problem is becoming increasingly prominent, and traditional fuel energy is decreasing day by day. The development and utilization of green and environmentally friendly energy is becoming more and more important. Photovoltaic power generation, as a new type of renewable energy, has the advantages of safety, reliability, cleanliness, pollution-free and renewable. Governments of various countries have also introduced policies to focus on supporting the photovoltaic industry. Photovoltaic power generation is a technology that uses the photovoltaic effect of the semiconductor interface to directly convert light energy into electrical energy. It is mainly composed of three parts: solar panels (modules), controllers and inverters, and the main components are composed of electronic components. After the solar cells are connected in series and packaged for protection, they can form a large-area solar cell module, which is combined with power controllers and other components to form a photovoltaic power generation device.

[0003] In the prior art, distributed photovoltaic power stations generally dispatch technicians to the site to conduct routine inspections of photovoltaic equipment at pre-set fixed periodic intervals. They are unable to conduct in-depth analysis of the operating status of photovoltaic equipment in distributed photovoltaic power stations, and when a power station fails, they are unable to quickly determine the failure of the photovoltaic equipment. Operation and maintenance personnel are unable to quickly handle the photovoltaic equipment in a timely manner according to the type of failure, resulting in low operation and maintenance efficiency.

[0004] In order to solve this problem, we propose an intelligent management method and system for distributed photovoltaic power stations based on big data monitoring. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a distributed photovoltaic power station intelligent management method and system based on big data monitoring, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is: An intelligent management method for a distributed photovoltaic power station based on big data monitoring, wherein the distributed photovoltaic power station includes at least one photovoltaic inverter system, and the photovoltaic inverter system includes a plurality of photovoltaic devices. The management method includes: Construct a distributed photovoltaic power station big data platform, which receives data writing requests from photovoltaic devices, which carry basic information, historical operating status data, and historical operation and maintenance data of various photovoltaic devices, and classifies and organizes the data; Based on the collected historical operating status data, the key features of the historical operating status of various photovoltaic equipment are extracted and in-depth statistical analysis is performed. Based on the in-depth statistical analysis results, the operating status definition model of various photovoltaic equipment is constructed; Based on the collected historical operation and maintenance data, a fault type assessment model for various photovoltaic equipment is constructed; Based on the basic information of each photovoltaic equipment and historical operation and maintenance data, formulate and implement the operation and maintenance plan for the first operation and maintenance cycle: Monitor and collect real-time operating status data of each photovoltaic device in real time, and determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model, recorded as the suspected fault type of the photovoltaic device, and a corresponding inspection plan is formulated based on the obtained suspected fault type; During the execution of the inspection plan, the operating status data of the photovoltaic equipment is continuously monitored and recorded as operating fault data; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data. Combined with the historical operation and maintenance data, the operation and maintenance plan of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. The first operation and maintenance cycle represents the current execution cycle, while the second operation and maintenance cycle refers to the next execution cycle that is about to begin.

[0007] Preferably, the preprocessing is performed based on the collected historical operation status data to extract key features of the historical operation status of each photovoltaic device, specifically including: Acquire historical operation status data of the photovoltaic device during the power-on and standby periods, wherein the historical operation status data includes at least the following parameters: device status information, device temperature data, device power consumption data, and device vibration information; Obtain several parameters in the historical operation status data as detection parameters, correspond the obtained detection parameters to a data sequence, and calculate the correlation between any two detection parameters based on the data sequence; The calculation method of the correlation between any two detection parameters is as follows: obtaining the frequency of simultaneous failure of any two detection parameters in a historical period, performing straight line fitting on each data sequence, obtaining the slope of the straight line corresponding to each data sequence, calculating the dispersion coefficient corresponding to each data sequence, calculating the Pearson correlation coefficient of any two data sequences, and calculating the correlation between any two detection parameters according to the frequency, slope, dispersion coefficient and Pearson correlation coefficient; The correlation calculation formula is: ; In the formula, is the correlation between the i-th detection parameter and the j-th detection parameter, is the frequency of abnormal occurrence of the i-th detection parameter and the j-th detection parameter at the same time in the historical period, is the Pearson correlation coefficient between the data sequence corresponding to the i-th detection parameter and the data sequence corresponding to the j-th detection parameter, is the slope corresponding to the data sequence of the i-th detection parameter, is the slope of the straight line corresponding to the data sequence of the jth detection parameter, is the discrete coefficient corresponding to the data sequence of the i-th detection parameter, is the discrete coefficient corresponding to the jth detection parameter sequence, is a vector consisting of the data of the i-th detection parameter, is a vector consisting of the data of the jth detection parameter, for and The inner product of , are weight parameters, is the model parameter, e is the natural constant; A correlation matrix is ​​constructed, and a plurality of key detection parameters are screened out from a plurality of detection parameters according to the correlation matrix.

[0008] Preferably, the in-depth statistical analysis is performed, and based on the in-depth statistical analysis results, a definition model of the operating status of various photovoltaic devices is constructed, which specifically includes: Obtain the data corresponding to each detection moment of the key detection parameters within the set time period, construct a key detection parameter matrix and reduce its dimension, and construct a reconstructed data matrix; Randomly select an element in the reconstructed data matrix , calculate the element The absolute value of the difference between all elements, preset a difference threshold, and set the difference threshold Compare with multiple difference absolute values ​​respectively. If the difference absolute value is greater than the difference threshold , then it is recorded as 1; if the absolute value of the difference is greater than the difference threshold , then it is recorded as 0, and the above comparison results are added to obtain a sum, which is recorded as the first distribution index of the element; In the reconstructed data matrix, the elements Get the first distribution index greater than the element in the window area centered The element corresponding to the first distribution index , calculate the element With all elements obtained The corresponding absolute value of the difference, and the maximum absolute value of the difference is taken as the element The second distribution index ; The size of the window area is n×n, where n is greater than or equal to 3; Calculate the state operation index corresponding to each element according to the first distribution index and the second distribution index corresponding to each element; The state operation indicators are: ; In the formula, Indicates the status of the running indicator. is the first distribution index, is the second distribution index, , is the weight coefficient, is a natural constant.

[0009] An indicator threshold is set, and the size of the operating status indicator is compared with the indicator threshold. When the operating status indicator is less than the indicator threshold, the operating status of the photovoltaic device is abnormal and deviates from the normal operating status mode; when the operating status indicator is greater than the indicator threshold, the operating status of the photovoltaic device is normal and does not deviate from the normal operating status mode.

[0010] Preferably, the fault type assessment model of various photovoltaic equipment is constructed based on the collected historical operation and maintenance data, specifically including: According to the historical operation and maintenance data of photovoltaic equipment, the historical operation fault outlier data is obtained based on box plot detection; The historical operation and maintenance data includes fault type information, fault frequency information and operation fault data; Analyze the outlier data of historical operation failures and obtain the outlier data difference index; Based on data analysis requirements, obtain the outlier data difference index threshold; According to the outlier data difference index and the outlier data difference index threshold, determine whether the historical operation fault outlier data is operation fault abnormal data; If the outlier data difference index exceeds the outlier data difference index threshold, the historical operation fault outlier data is erroneous data and the data is removed; If the outlier data difference index does not exceed the outlier data difference index threshold, the historical operation fault outlier data is available; Match the above historical operation fault outlier data with the corresponding fault type information, obtain the fault waveform of the historical operation fault outlier data, and merge the data to obtain a data set; Based on the fault waveform and fault type information of historical operating fault outlier data, a convolutional neural network model is established for training to obtain a fault type assessment model. When evaluating the operation faults of each photovoltaic device, the operation fault data is input into the fault type evaluation model, and the fault waveform is detected and compared by the fault type evaluation model to determine at least one fault type; The outlier data difference index is: ; In the formula, is the outlier data difference index of the x-th historical running fault outlier data, For the xth historical run fault outlier data, is the ith historical running fault outlier data of the fault type described by the xth historical running fault outlier data, is the i+1th historical operation fault outlier data of the fault type described in the xth historical operation fault outlier data, is the interval time between the i-th historical operation fault outlier data and the i+1-th historical operation fault outlier data of the fault type described in the x-th historical operation fault outlier data, and n is the total number of historical operation fault outlier data of the fault type described in the x-th historical operation fault outlier data.

[0011] Preferably, the formulating and executing the operation and maintenance plan for the first operation and maintenance cycle based on the basic information and historical operation and maintenance data of each photovoltaic device specifically includes: The operation and maintenance plan of the first operation and maintenance cycle includes: an execution phase and an operation and maintenance cycle interval setting; The execution stage includes: monitoring and collecting the operation status data of the photovoltaic equipment in real time through the SCADA system, and processing the collected data using a real-time data analysis tool, which is recorded as real-time operation status data; Determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operation status mode does not deviate from normal, the data is uploaded to the historical operation status data; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model and recorded as a suspected fault type of the photovoltaic device; Develop a corresponding inspection plan based on the suspected fault type obtained; The inspection plan at least includes inspection objectives, inspection time, inspection content and personnel arrangements; During the execution of the inspection plan, the photovoltaic equipment is continuously monitored, and the equipment status information, equipment temperature data, equipment power consumption data and equipment vibration information of the photovoltaic equipment are recorded and recorded as operation fault data; If no fault is found during the execution of the inspection plan, the operation fault data is considered as no-fault data and the data is removed; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data of the corresponding photovoltaic equipment. Combined with the historical operation and maintenance data, the execution phase of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. After the execution phase is over, the operation and maintenance cycle interval of the photovoltaic equipment is set. After the operation and maintenance cycle interval is set, it enters a waiting state until the first operation and maintenance cycle is over. After the first operation and maintenance cycle is over, the operation and maintenance plan of the second operation and maintenance cycle is immediately executed.

[0012] Preferably, the operation and maintenance cycle interval setting specifically includes: According to the basic information of various photovoltaic equipment, the basic information of the photovoltaic equipment includes equipment model, equipment location information, equipment specification information and equipment configuration information, the initial operation and maintenance cycle interval is set; Combine the historical operation and maintenance data of each photovoltaic device to analyze the fault type and frequency information of each photovoltaic device; Shorten the initial operation and maintenance cycle interval for adjustment based on PV equipment with higher failure frequency; Keep the initial operation and maintenance cycle interval unchanged based on the photovoltaic equipment with low failure frequency; The reset operation and maintenance cycle interval duration is obtained, and the operation and maintenance cycle interval duration is updated and applied to the first operation and maintenance cycle, and the adjusted operation and maintenance plan becomes the guiding basis for the second operation and maintenance cycle, thereby achieving continuous optimization of operation and maintenance management.

[0013] Furthermore, the present invention also proposes a distributed photovoltaic power station intelligent management system based on big data monitoring, which is used to implement any of the above management methods, including: A data acquisition module, which is used to obtain basic information, historical operating status data, and historical operation and maintenance data of each photovoltaic device, classify and sort them, and then transmit them to the big data platform module; The big data platform module is used to receive the write request from the data acquisition module and store the classified data into the distributed file database; An operation and maintenance plan execution module is configured to formulate a detailed operation and maintenance plan based on the historical operation and maintenance data of the big data platform module, monitor the real-time operation status data of the photovoltaic equipment in real time during the execution of the operation and maintenance plan, and determine whether the collected real-time operation status data deviates from the normal operation status mode. If the operation status mode does not deviate from normal, the data is uploaded to the data acquisition module; if the operation status mode deviates from normal, the suspected fault type of the photovoltaic equipment is determined, and a detailed inspection plan is formulated based on the suspected fault type. The data obtained in the inspection plan is uploaded to the data acquisition module, and dynamic adjustment of the operation and maintenance plan for the next cycle is achieved based on the big data platform module.

[0014] Preferably, the data acquisition module specifically includes: A data acquisition unit, the data acquisition unit is used to acquire basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device; A data classification processing unit, wherein the data classification processing unit is used to classify and process the historical operating status data and historical operation and maintenance data of each photovoltaic device according to the basic information of each photovoltaic device; A data transmission unit, wherein the data transmission unit is used to send a write request to the big data module and transmit the classified and sorted data to the big data platform module.

[0015] Preferably, the big data platform module specifically includes: A data receiving unit, the data receiving unit is used to receive a write request from the data transmission unit; The data storage unit is used to store the basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device after classification into a distributed file database, providing a basis for data analysis and application of the operation and maintenance plan execution module.

[0016] Preferably, the operation and maintenance plan execution module specifically includes: An operation and maintenance plan formulation unit, the operation and maintenance plan formulation unit is used to formulate an operation and maintenance plan based on historical operation and maintenance data of the big data platform module; A real-time monitoring unit, wherein the real-time monitoring unit is used to monitor the real-time operating status data of the photovoltaic device in real time; An operation status definition unit, which is used to construct an operation status definition model for various photovoltaic devices based on the historical operation and maintenance data of the big data platform module, and to determine whether the collected real-time operation status data deviates from a normal operation status mode based on the operation status definition model by real-time monitoring of the operation status data of the photovoltaic devices; An operation data uploading unit, wherein the operation data uploading unit is used to upload the real-time operation status data of the photovoltaic device detected by the operation status definition unit and in the normal operation state mode monitored by the real-time monitoring unit to the historical operation status data of the data acquisition unit; A fault type assessment unit, which is used to obtain historical operation and maintenance data and build a fault type assessment model for various types of photovoltaic equipment, assess the fault type of the photovoltaic equipment that has been detected by the operation state definition unit and deviates from the normal operation state mode, and record it as a suspected fault type; An inspection plan formulation unit, the inspection plan formulation unit is used to formulate a detailed inspection plan according to the suspected fault type obtained by the fault type assessment unit, the inspection plan at least including: inspection time, inspection location, inspection items, required tools and materials; A fault data uploading unit, the fault data uploading unit is used to upload the fault type information and operation fault data of the photovoltaic equipment determined in the inspection plan to the historical operation and maintenance data in the data acquisition unit; An operation and maintenance cycle setting unit, which is used to set the initial operation and maintenance cycle interval according to the basic information of the photovoltaic equipment, and dynamically adjust the operation and maintenance cycle interval in combination with historical operation and maintenance data; The data updating unit is used to upload the data collected in the current operation and maintenance plan to the data acquisition unit, so that the operation status definition unit, the fault type assessment unit, and the operation and maintenance plan formulation unit can adapt to the actual conditions of the photovoltaic equipment and make adjustments.

[0017] Compared with the prior art, the present invention provides a distributed photovoltaic power station intelligent management method and system based on big data monitoring, which has the following beneficial effects: 1. This scheme proposes a distributed photovoltaic power station intelligent management method and system based on big data monitoring. By constructing a distributed photovoltaic power station big data platform, the system monitors and collects the operating status data of the photovoltaic equipment in real time during the execution of the photovoltaic equipment operation and maintenance plan, and uses the operation status definition model to analyze the collected real-time operation status. The abnormal data obtained is analyzed by the fault type assessment model to obtain the suspected fault type, and the corresponding inspection plan is formulated based on the suspected fault type. This management method can analyze the operating status of the equipment during the execution of the operation and maintenance plan, identify potential faults in advance, and detect abnormalities in time before the fault occurs. The fault type is diagnosed based on the abnormal data of the equipment. The suspected fault type determined by the fault type assessment model can locate the problem more accurately, avoid blind inspections, and improve the pertinence of maintenance work.

[0018] 2. This scheme proposes a distributed photovoltaic power station intelligent management method and system based on big data monitoring. During the execution of the operation and maintenance plan, the real-time operation status data of the photovoltaic equipment monitored and collected by the system in real time is analyzed according to the operation status definition model to see whether it deviates from the normal operation status mode. If the operation status mode does not deviate from normal, the data is uploaded to the historical operation status data of the big data platform; if the operation status mode deviates from normal, the suspected fault type of the photovoltaic equipment is determined according to the fault type assessment model. When the fault type of the photovoltaic equipment is determined in the inspection plan, the determined fault type information and the operation fault data are uploaded to the historical operation and maintenance data of the big data platform. The operation status assessment model and the fault type assessment model of the photovoltaic equipment are updated in real time based on the data in the big data platform, which is closer to the actual operation status of the equipment and improves the adaptability and accuracy of the model.

[0019] 3. This plan proposes an intelligent management method and system for distributed photovoltaic power stations based on big data monitoring. This management method can reasonably set the operation and maintenance cycle interval according to the basic information of the equipment and historical operation and maintenance data. For equipment with a high failure frequency, the operation and maintenance cycle interval can be shortened to timely discover and handle potential failures, ensure that photovoltaic equipment is maintained in a timely manner, and ensure its reliable operation. During the operation and maintenance period, the big data platform will update the data in real time, and the operation and maintenance cycle interval can be dynamically adjusted according to the historical operation and maintenance data to ensure that the operation and maintenance strategy always adapts to the actual situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of the intelligent management method of distributed photovoltaic power stations based on big data monitoring proposed by the present invention.

[0021] Figure 2 This is a flow chart of constructing an operating status definition model for various photovoltaic devices in the present invention.

[0022] Figure 3 This is a flow chart of constructing a fault type assessment model for various photovoltaic devices in the present invention.

[0023] Figure 4 This is a framework diagram of the distributed photovoltaic power station intelligent management system based on big data monitoring proposed in the present invention. DETAILED DESCRIPTION

[0024] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention will be further described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] In order to address the shortcomings of the existing technology, such as Figure 1-4 As shown, the present invention provides a distributed photovoltaic power station intelligent management method and system based on big data monitoring, the distributed photovoltaic power station includes at least one photovoltaic inverter system, the photovoltaic inverter system includes multiple photovoltaic devices, and the management method includes: Construct a distributed photovoltaic power station big data platform, which receives data writing requests from photovoltaic devices, which carry basic information, historical operating status data, and historical operation and maintenance data of various photovoltaic devices, and classifies and organizes the data; Based on the collected historical operating status data, the key features of the historical operating status of various photovoltaic equipment are extracted and in-depth statistical analysis is performed. Based on the in-depth statistical analysis results, the operating status definition model of various photovoltaic equipment is constructed; Based on the collected historical operation and maintenance data, a fault type assessment model for various photovoltaic equipment is constructed; Based on the basic information of each photovoltaic equipment and historical operation and maintenance data, formulate and implement the operation and maintenance plan for the first operation and maintenance cycle: Monitor and collect real-time operating status data of each photovoltaic device in real time, and determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model, recorded as the suspected fault type of the photovoltaic device, and a corresponding inspection plan is formulated based on the obtained suspected fault type; During the execution of the inspection plan, the operating status data of the photovoltaic equipment is continuously monitored and recorded as operating fault data; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data. Combined with the historical operation and maintenance data, the operation and maintenance plan of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. The first operation and maintenance cycle represents the current execution cycle, while the second operation and maintenance cycle refers to the next execution cycle that is about to begin; It should be noted that a distributed photovoltaic power station big data platform is constructed. The big data platform receives data write requests from photovoltaic devices. The data write requests carry basic information, historical operating status data, and historical operation and maintenance data of various photovoltaic devices, and classifies and organizes the data. Specifically, it includes: collecting various types of data from photovoltaic devices, including basic information, historical operating status data, and historical operation and maintenance data, ensuring that the historical operating status data and historical operation and maintenance data are accurately timestamped when collected. The timestamp includes date and time. The data is transmitted to the big data platform through a data transmission protocol for data formatting so that all data follows a unified format standard. The MongoDB database is used to manage and store data. The data is further mapped and subdivided according to the basic information of the photovoltaic equipment. The data is stored in the form of documents. Each document contains at least one field. Each field has a clear definition and format. The timestamp is used as a key field, and an index is established for the field to retrieve and filter the corresponding data for each time period, which is convenient for management and query.

[0026] Furthermore, based on the collected historical operating status data, preprocessing is performed to extract the key features of the historical operating status of each photovoltaic device, including: Acquire historical operation status data of the photovoltaic equipment during the power-on and standby periods, the historical operation status data including at least the following parameters: equipment status information, equipment temperature data, equipment power consumption data, and equipment vibration information; Obtain several parameters in the historical operation status data as detection parameters, correspond the obtained detection parameters to a data sequence, and calculate the correlation between any two detection parameters based on the data sequence; The calculation method of the correlation between any two detection parameters is as follows: obtain the frequency of simultaneous failure of any two detection parameters in the historical period, perform straight line fitting on each data series, obtain the slope of the straight line corresponding to each data series, calculate the dispersion coefficient corresponding to each data series, calculate the Pearson correlation coefficient of any two data series, and calculate the correlation between any two detection parameters based on the frequency, slope, dispersion coefficient and Pearson correlation coefficient; A correlation matrix is ​​constructed based on the correlation, and multiple key detection parameters are screened out from several detection parameters according to the correlation matrix; the correlation calculation formula is: ; In the formula, is the correlation between the i-th detection parameter and the j-th detection parameter, is the frequency of abnormal occurrence of the i-th detection parameter and the j-th detection parameter at the same time in the historical period, is the Pearson correlation coefficient between the data sequence corresponding to the i-th detection parameter and the data sequence corresponding to the j-th detection parameter, is the slope corresponding to the data sequence of the i-th detection parameter, is the slope of the straight line corresponding to the data sequence of the jth detection parameter, is the discrete coefficient corresponding to the data sequence of the i-th detection parameter, is the discrete coefficient corresponding to the jth detection parameter sequence, is a vector consisting of the data of the i-th detection parameter, is a vector consisting of the data of the jth detection parameter, for and The inner product of , are weight parameters, is the model parameter and e is a natural constant.

[0027] Specifically, each element in the correlation matrix represents the correlation between the ith and jth detection parameters. A threshold is set. When the correlation between the two parameters is greater than or equal to the threshold, the correlation between the two parameters is stronger; when the correlation between the two parameters is less than the threshold, the correlation between the two parameters is weaker. According to the correlation matrix, the two parameters with the strongest correlation are selected as key detection parameters.

[0028] Furthermore, an in-depth statistical analysis is conducted, and based on the results of the in-depth statistical analysis, a definition model for the operating status of various photovoltaic equipment is constructed, including: Obtain the data corresponding to each detection moment of the key detection parameters within the set time period, construct a key detection parameter matrix and reduce its dimension, and construct a reconstructed data matrix; Randomly select an element in the reconstructed data matrix , calculate the element The absolute value of the difference between all elements, preset a difference threshold, and set the difference threshold Compare with multiple difference absolute values ​​respectively. If the difference absolute value is greater than the difference threshold , then it is recorded as 1; if the absolute value of the difference is greater than the difference threshold , then it is recorded as 0, and the above comparison results are added to obtain a sum, which is recorded as the first distribution index of the element; In the reconstructed data matrix, the elements Get the first distribution index greater than the element in the window area centered The element corresponding to the first distribution index , calculate the element With all elements obtained The corresponding absolute value of the difference, and the maximum absolute value of the difference is taken as the element The second distribution index ; The size of the window area is n×n, where n is greater than or equal to 3; Calculate the state operation index corresponding to each element according to the first distribution index and the second distribution index corresponding to each element; The status operation indicators are: ; In the formula, Indicates the status of the running indicator. is the first distribution index, is the second distribution index, , is the weight coefficient, is a natural constant.

[0029] An indicator threshold is set, and the size of the operating status indicator is compared with the indicator threshold. When the operating status indicator is less than the indicator threshold, the operating status of the photovoltaic equipment is abnormal and deviates from the normal operating status mode; when the operating status indicator is greater than the indicator threshold, the operating status of the photovoltaic equipment is normal and does not deviate from the normal operating status mode.

[0030] Furthermore, based on the collected historical operation and maintenance data, a fault type assessment model for various photovoltaic equipment is constructed, including: According to the historical operation and maintenance data of photovoltaic equipment, the historical operation fault outlier data is obtained based on box plot detection; Historical operation and maintenance data includes fault type information, fault frequency information, and operational fault data; Analyze the outlier data of historical operation failures and obtain the outlier data difference index; Based on data analysis requirements, obtain the outlier data difference index threshold; According to the outlier data difference index and the outlier data difference index threshold, determine whether the historical operation fault outlier data is operation fault abnormal data; If the outlier data difference index exceeds the outlier data difference index threshold, the historical operation fault outlier data is erroneous data and the data is removed; If the outlier data difference index does not exceed the outlier data difference index threshold, the historical operation fault outlier data is available; Match the above historical operation fault outlier data with the corresponding fault type information, obtain the fault waveform of the historical operation fault outlier data, and merge the data to obtain a data set; Based on the fault waveform and fault type information of historical operating fault outlier data, a convolutional neural network model is established for training to obtain a fault type assessment model. When evaluating the operation faults of each photovoltaic device, the operation fault data is input into the fault type evaluation model, and the fault waveform is detected and compared by the fault type evaluation model to determine at least one fault type; The outlier data difference index is: ; In the formula, is the outlier data difference index of the x-th historical running fault outlier data, For the xth historical run fault outlier data, is the ith historical running fault outlier data of the fault type x, is the i+1th historical running fault outlier data of the xth historical running fault outlier data fault type, is the interval time between the i-th historical operation fault outlier data and the i+1-th historical operation fault outlier data of the x-th historical operation fault outlier data fault type, and n is the total number of historical operation fault outlier data of the x-th historical operation fault outlier data fault type.

[0031] Specifically, the fault waveforms extracted from the historical operation and maintenance data are time series data, including but not limited to curves of voltage, current, temperature and other parameters changing over time, and one or more fault type labels are assigned to each fault waveform to represent the type of fault corresponding to the waveform; Normalize the fault waveform data to the range of [0, 1] to improve the stability and convergence speed of model training. Increase data diversity through rotation, translation, and scaling to improve the generalization ability of the model. Input the normalized data as the input feature map into the convolutional neural network model. For [0, 1] normalization, the following formula is used: ; In the formula, is the value in the fault waveform data, is the normalized value, is the minimum value in the fault waveform data, is the maximum value in the fault waveform data; The convolutional neural network model includes: The convolution layer uses convolution kernels to extract local features of the fault waveform, including: Slide the convolution kernel on the input feature map, calculate the dot product between the convolution kernel and the local area of ​​the input feature map, and obtain each pixel value of the output feature map (convolution result). Set the input feature map size to , and the convolution kernel size to , then the calculation formula of the convolution operation is: ; In the formula, is the output feature map at position The pixel value of is the convolution kernel at position The weight of is the pixel value of the input feature map in the area covered by the convolution kernel, is the bias term; The activation layer uses the ReLU function to increase nonlinearity, and its calculation formula is: ; The ReLU function simply sets all negative values ​​to 0 and leaves positive values ​​unchanged; The pooling layer reduces the spatial dimension of the feature map and extracts the main features. The calculation formula is: ; In the formula, The pooling layer is at position The pixel value of and is the index set of the pooling window in the horizontal and vertical directions, is the pixel value of the input pooling window; Fully connected layer, mapping the extracted features to the final fault type; The output layer uses the softmax function to output the probability distribution of the fault type.

[0032] Furthermore, based on the basic information of each photovoltaic device and historical operation and maintenance data, an operation and maintenance plan for the first operation and maintenance cycle is formulated and implemented, including: The operation and maintenance plan for the first operation and maintenance cycle includes: the execution phase and the setting of the interval between operation and maintenance cycles; The execution phase includes: real-time monitoring and collection of operating status data of photovoltaic equipment through the SCADA system, and processing the collected data using real-time data analysis tools, which are recorded as real-time operating status data; Determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operation status mode does not deviate from normal, the data is uploaded to the historical operation status data; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model and recorded as a suspected fault type of the photovoltaic device; Develop a corresponding inspection plan based on the suspected fault type obtained; The inspection plan shall at least include inspection objectives, inspection time, inspection content and personnel arrangements; During the execution of the inspection plan, the photovoltaic equipment is continuously monitored, and the equipment status information, equipment temperature data, equipment power consumption data and equipment vibration information of the photovoltaic equipment are recorded and recorded as operation fault data; If no fault is found during the execution of the inspection plan, the operation fault data is considered as no-fault data and the data is removed; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data of the corresponding photovoltaic equipment. Combined with the historical operation and maintenance data, the execution phase of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. After the execution phase is over, the operation and maintenance cycle interval of the photovoltaic equipment is set. After the operation and maintenance cycle interval is set, it enters a waiting state until the first operation and maintenance cycle is over. After the first operation and maintenance cycle is over, the operation and maintenance plan of the second operation and maintenance cycle is immediately executed.

[0033] Furthermore, the operation and maintenance cycle interval is set, including: According to the basic information of various photovoltaic equipment, including equipment model, equipment location information, equipment specification information and equipment configuration information, the initial operation and maintenance cycle interval is set; Combine the historical operation and maintenance data of each photovoltaic device to analyze the fault type and frequency information of each photovoltaic device; Shorten the initial operation and maintenance cycle interval for adjustment based on PV equipment with higher failure frequency; Keep the initial operation and maintenance cycle interval unchanged based on the photovoltaic equipment with low failure frequency; The reset operation and maintenance cycle interval is used in the first operation and maintenance cycle, and the fault type information and fault frequency information in the operation and maintenance cycle are uploaded to the big data platform for update and application. After adjustment, it becomes the guiding basis for the second operation and maintenance cycle, thereby realizing continuous optimization of operation and maintenance management.

[0034] Specifically, for device collections , based on each photovoltaic device Basic information to determine the initial operation and maintenance cycle interval of the equipment , based on the device The fault type information and fault frequency information of photovoltaic equipment The operation and maintenance cycle interval is adjusted according to the following calculation formula: ; In the formula, is the failure frequency of the equipment, that is, the number of failures during the operating time divided by the operating time, is the minimum maintenance cycle interval set. is the initial operation and maintenance cycle interval, For the The frequency of each fault type, For the The weight of each fault type; It should be noted that the above formula calculates the adjustment factor based on the fault frequency and the fault type weight, and adjusts the operation and maintenance cycle interval according to the factor. Finally, by taking the maximum and minimum values, it is ensured that the final operation and maintenance cycle interval is within the practical range.

[0035] Furthermore, the present invention also proposes a distributed photovoltaic power station intelligent management system based on big data monitoring, which is used to implement any of the above management methods, including: Data collection module: The data collection module is used to obtain basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device, classify and sort them, and then transmit them to the big data platform module; The big data platform module is used to receive the write request from the data acquisition module and store the classified data into the distributed file database; The operation and maintenance plan execution module formulates a detailed operation and maintenance plan based on the historical operation and maintenance data of the big data platform module. During the execution of the operation and maintenance plan, the real-time operation status data of the photovoltaic equipment is monitored in real time to determine whether the collected real-time operation status data deviates from the normal operation status mode. If the operation status mode does not deviate from normal, the data is uploaded to the data acquisition module; if the operation status mode deviates from normal, the suspected fault type of the photovoltaic equipment is determined, and a detailed inspection plan is formulated based on the suspected fault type. The data obtained in the inspection plan is uploaded to the data acquisition module, and the dynamic adjustment of the operation and maintenance plan for the next cycle is realized based on the big data platform module.

[0036] Furthermore, the data acquisition module specifically includes: A data acquisition unit, which is used to acquire basic information, historical operating status data, and historical operation and maintenance data of each photovoltaic device; A data classification processing unit, which is used to classify and process the historical operating status data and historical operation and maintenance data of each photovoltaic device according to the basic information of each photovoltaic device; The data transmission unit is used to send write requests to the big data module and transmit the classified and sorted data to the big data platform module.

[0037] Furthermore, the big data platform module specifically includes: A data receiving unit, the data receiving unit is used to receive a write request from the data transmission unit; The data storage unit is used to store the basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device after classification into a distributed file database, providing a basis for data analysis and application of the operation and maintenance plan execution module.

[0038] Furthermore, the operation and maintenance plan execution module specifically includes: An operation and maintenance plan formulation unit, which is used to formulate an operation and maintenance plan based on the historical operation and maintenance data of the big data platform module; A real-time monitoring unit, which is used to monitor the real-time operating status data of the photovoltaic equipment in real time; The operation status definition unit is used to build an operation status definition model for various photovoltaic equipment based on the historical operation and maintenance data of the big data platform module, and to determine whether the collected real-time operation status data deviates from the normal operation status mode based on the operation status definition model by monitoring the operation status data of the photovoltaic equipment in real time; An operation data uploading unit, the operation data uploading unit is used to upload the real-time operation status data of the photovoltaic equipment detected by the operation status definition unit and in the normal operation state mode monitored by the real-time monitoring unit to the historical operation status data of the data acquisition unit; A fault type assessment unit is used to obtain historical operation and maintenance data and build a fault type assessment model for various photovoltaic equipment, and to assess the fault type of the photovoltaic equipment that has been detected by the operation state definition unit and deviates from the normal operation state mode, and record it as a suspected fault type; An inspection plan formulation unit is used to formulate a detailed inspection plan according to the suspected fault type obtained by the fault type assessment unit. The inspection plan at least includes: inspection time, inspection location, inspection items, and required tools and materials; A fault data uploading unit, which is used to upload the fault type information and operation fault data of the photovoltaic equipment determined in the inspection plan to the historical operation and maintenance data in the data acquisition unit; The operation and maintenance cycle setting unit is used to set the initial operation and maintenance cycle interval according to the basic information of the photovoltaic equipment, and dynamically adjust the operation and maintenance cycle interval in combination with historical operation and maintenance data; The data updating unit is used to upload the data collected in the current operation and maintenance plan to the data acquisition unit, so that the operation status definition unit, the fault type assessment unit, and the operation and maintenance plan formulation unit can adapt to the actual conditions of the photovoltaic equipment and make adjustments.

[0039] In summary, the advantages of the present invention are: by constructing a distributed photovoltaic power station big data platform, in the process of executing the photovoltaic equipment operation and maintenance plan, the system monitors and collects the operating status data of the photovoltaic equipment in real time, adopts the operating status definition model to analyze the collected real-time operating status, and obtains the suspected fault type after analyzing the obtained abnormal data through the fault type evaluation model, and formulates a corresponding inspection plan based on the obtained suspected fault type. The management method can analyze the operating status of the equipment during the execution of the operation and maintenance plan, identify potential faults in advance, and find abnormalities in time before the fault occurs, and diagnose the fault type based on the abnormal data of the equipment. The suspected fault type determined by the fault type evaluation model can more accurately locate the problem, avoid blind inspections, and improve the pertinence of maintenance work; During the execution of the operation and maintenance plan, the real-time operation status data of the photovoltaic equipment monitored and collected by the operation status definition model analysis system in real time is analyzed to see whether it deviates from the normal operation status mode. If the operation status mode does not deviate from the normal, the data is uploaded to the historical operation status data of the big data platform; if the operation status mode deviates from the normal, the suspected fault type of the photovoltaic equipment is determined according to the fault type assessment model. When the fault type of the photovoltaic equipment is determined in the inspection plan, the determined fault type information and the operation fault data are uploaded to the historical operation and maintenance data of the big data platform. The operation status assessment model and the fault type assessment model of the photovoltaic equipment are updated in real time based on the data in the big data platform, which are closer to the actual operation status of the equipment and improve the adaptability and accuracy of the model. This management method can reasonably set the operation and maintenance cycle interval according to the basic information of the equipment and historical operation and maintenance data. For equipment with a high failure frequency, the operation and maintenance cycle interval can be shortened to promptly discover and handle potential failures, ensure that photovoltaic equipment is maintained in a timely manner, and guarantee its reliable operation. During the operation and maintenance period, the big data platform will update the data in real time, and the operation and maintenance cycle interval can be dynamically adjusted according to historical operation and maintenance data to ensure that the operation and maintenance strategy always adapts to the actual situation.

[0040] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A distributed photovoltaic power station intelligent management method based on big data monitoring, wherein the distributed photovoltaic power station comprises at least one photovoltaic inverter system, and the photovoltaic inverter system comprises a plurality of photovoltaic devices. It is characterized in that The management method comprises: Construct a distributed photovoltaic power station big data platform, which receives data writing requests from photovoltaic devices, which carry basic information, historical operating status data, and historical operation and maintenance data of various photovoltaic devices, and classifies and organizes the data; Based on the collected historical operating status data, the key features of the historical operating status of various photovoltaic equipment are extracted and in-depth statistical analysis is performed. Based on the in-depth statistical analysis results, the operating status definition model of various photovoltaic equipment is constructed; Based on the collected historical operation and maintenance data, a fault type assessment model for various photovoltaic equipment is constructed; Based on the basic information of each photovoltaic equipment and historical operation and maintenance data, formulate and implement the operation and maintenance plan for the first operation and maintenance cycle: Monitor and collect real-time operating status data of each photovoltaic device in real time, and determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model, recorded as the suspected fault type of the photovoltaic device, and a corresponding inspection plan is formulated based on the obtained suspected fault type; During the execution of the inspection plan, the operating status data of the photovoltaic equipment is continuously monitored and recorded as operating fault data; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data. Combined with the historical operation and maintenance data, the operation and maintenance plan of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. The first operation and maintenance cycle represents the current execution cycle, while the second operation and maintenance cycle refers to the next execution cycle that is about to begin.

2. According to claim 1, the intelligent management method of distributed photovoltaic power station based on big data monitoring, It is characterized in that The preprocessing is performed based on the collected historical operation status data to extract key features of the historical operation status of each photovoltaic device, specifically including: Acquire historical operation status data of the photovoltaic device during the power-on and standby periods, wherein the historical operation status data includes at least the following parameters: device status information, device temperature data, device power consumption data, and device vibration information; Obtain several parameters in the historical operation status data as detection parameters, correspond the obtained detection parameters to a data sequence, and calculate the correlation between any two detection parameters based on the data sequence; The calculation method of the correlation between any two detection parameters is as follows: obtaining the frequency of simultaneous failure of any two detection parameters in a historical period, performing straight line fitting on each data sequence, obtaining the slope of the straight line corresponding to each data sequence, calculating the dispersion coefficient corresponding to each data sequence, calculating the Pearson correlation coefficient of any two data sequences, and calculating the correlation between any two detection parameters according to the frequency, slope, dispersion coefficient and Pearson correlation coefficient; Construct a correlation matrix, and select multiple key detection parameters from a number of detection parameters according to the correlation matrix; The correlation calculation formula is: ； In the formula, is the correlation between the i-th detection parameter and the j-th detection parameter, is the frequency of abnormal occurrence of the i-th detection parameter and the j-th detection parameter at the same time in the historical period, is the Pearson correlation coefficient between the data sequence corresponding to the i-th detection parameter and the data sequence corresponding to the j-th detection parameter, is the slope corresponding to the data sequence of the i-th detection parameter, is the slope of the straight line corresponding to the data sequence of the jth detection parameter, is the discrete coefficient corresponding to the data sequence of the i-th detection parameter, is the discrete coefficient corresponding to the jth detection parameter sequence, is a vector consisting of the data of the i-th detection parameter, is a vector consisting of the data of the jth detection parameter, for and The inner product of , are weight parameters, is the model parameter and e is a natural constant.

3. According to claim 2, the intelligent management method of distributed photovoltaic power station based on big data monitoring, It is characterized in that The in-depth statistical analysis is performed, and based on the in-depth statistical analysis results, the operation status definition model of various photovoltaic equipment is constructed, specifically including: Obtain the data corresponding to each detection moment of the key detection parameters within the set time period, construct a key detection parameter matrix and reduce its dimension, and construct a reconstructed data matrix; Randomly select an element in the reconstructed data matrix , calculate the element The absolute value of the difference between all elements, preset a difference threshold, and set the difference threshold Compare with multiple difference absolute values ​​respectively. If the difference absolute value is greater than the difference threshold , then it is recorded as 1; if the absolute value of the difference is greater than the difference threshold , then it is recorded as 0, and the above comparison results are added to obtain a sum, which is recorded as the first distribution index of the element; In the reconstructed data matrix, the elements Get the first distribution index greater than the element in the window area centered The element corresponding to the first distribution index , calculate the element With all elements obtained The corresponding absolute value of the difference, and the maximum absolute value of the difference is taken as the element The second distribution index ; The size of the window area is n×n, where n is greater than or equal to 3; Calculate the state operation index corresponding to each element according to the first distribution index and the second distribution index corresponding to each element; An indicator threshold is set, and the size of the operating status indicator is compared with the indicator threshold. When the operating status indicator is less than the indicator threshold, the operating status of the photovoltaic device is abnormal and deviates from the normal operating status mode; when the operating status indicator is greater than the indicator threshold, the operating status of the photovoltaic device is normal and does not deviate from the normal operating status mode; The state operation indicators are: ； In the formula, Indicates the status of the running indicator. is the first distribution index, is the second distribution index, , is the weight coefficient, is a natural constant.

4. According to claim 3, the intelligent management method of distributed photovoltaic power station based on big data monitoring, It is characterized in that Based on the collected historical operation and maintenance data, a fault type assessment model for various photovoltaic equipment is constructed, specifically including: According to the historical operation and maintenance data of photovoltaic equipment, the historical operation fault outlier data is obtained based on box plot detection; The historical operation and maintenance data includes fault type information, fault frequency information and operation fault data; Analyze the outlier data of historical operation failures and obtain the outlier data difference index; Based on data analysis requirements, obtain the outlier data difference index threshold; According to the outlier data difference index and the outlier data difference index threshold, determine whether the historical operation fault outlier data is operation fault abnormal data; If the outlier data difference index exceeds the outlier data difference index threshold, the historical operation fault outlier data is erroneous data and the data is removed; If the outlier data difference index does not exceed the outlier data difference index threshold, the historical operation fault outlier data is available; Match the above historical operation fault outlier data with the corresponding fault type information, obtain the fault waveform of the historical operation fault outlier data, and merge the data to obtain a data set; Based on the fault waveform and fault type information of historical operating fault outlier data, a convolutional neural network model is established for training to obtain a fault type assessment model. When evaluating the operation faults of each photovoltaic device, the operation fault data is input into the fault type evaluation model, and the fault waveform is detected and compared by the fault type evaluation model to determine at least one fault type; The outlier data difference index is: ； In the formula, is the outlier data difference index of the x-th historical running fault outlier data, For the xth historical run fault outlier data, is the ith historical running fault outlier data of the fault type described by the xth historical running fault outlier data, is the i+1th historical operation fault outlier data of the fault type described in the xth historical operation fault outlier data, is the interval time between the i-th historical operation fault outlier data and the i+1-th historical operation fault outlier data of the fault type described in the x-th historical operation fault outlier data, and n is the total number of historical operation fault outlier data of the fault type described in the x-th historical operation fault outlier data.

5. According to claim 4, the intelligent management method of distributed photovoltaic power station based on big data monitoring, It is characterized in that The operation and maintenance plan for the first operation and maintenance cycle is formulated and executed based on the basic information of each photovoltaic device and the historical operation and maintenance data, specifically including: The operation and maintenance plan of the first operation and maintenance cycle includes: an execution phase and an operation and maintenance cycle interval setting; The execution stage includes: monitoring and collecting the operation status data of the photovoltaic equipment in real time through the SCADA system, and processing the collected data using a real-time data analysis tool, which is recorded as real-time operation status data; Determine whether the collected real-time operating status data deviates from the normal operating status mode according to the operating status definition model; If the operation status mode does not deviate from normal, the data is uploaded to the historical operation status data; If the operating state mode deviates from normal, at least one fault type is determined according to the fault type assessment model and recorded as a suspected fault type of the photovoltaic device; Develop a corresponding inspection plan based on the suspected fault type obtained; The inspection plan at least includes inspection objectives, inspection time, inspection content and personnel arrangements; During the execution of the inspection plan, the photovoltaic equipment is continuously monitored, and the equipment status information, equipment temperature data, equipment power consumption data and equipment vibration information of the photovoltaic equipment are recorded and recorded as operation fault data; If the inspection plan is executed, the running fault data is considered as no-fault data and the data is removed; If the fault type is determined during the execution of the inspection plan, the fault type information and operation fault data of the photovoltaic equipment will be uploaded to the historical operation and maintenance data of the corresponding photovoltaic equipment. Combined with the historical operation and maintenance data, the execution phase of the first operation and maintenance cycle will be updated and adjusted in a timely manner. The adjusted operation and maintenance plan will become the guiding basis for the second operation and maintenance cycle, thus achieving continuous optimization of operation and maintenance management. After the execution phase is over, the operation and maintenance cycle interval of the photovoltaic equipment is set. After the operation and maintenance cycle interval is set, it enters a waiting state until the first operation and maintenance cycle is over. After the first operation and maintenance cycle is over, the operation and maintenance plan of the second operation and maintenance cycle is immediately executed.

6. According to claim 5, the intelligent management method of distributed photovoltaic power station based on big data monitoring, It is characterized in that The operation and maintenance cycle interval setting specifically includes: According to the basic information of various photovoltaic equipment, the basic information of the photovoltaic equipment includes equipment model, equipment location information, equipment specification information and equipment configuration information, the initial operation and maintenance cycle interval is set; Combine the historical operation and maintenance data of each photovoltaic device to analyze the fault type and frequency information of each photovoltaic device; Shorten the initial operation and maintenance cycle interval for adjustment based on PV equipment with higher failure frequency; Keep the initial operation and maintenance cycle interval unchanged based on the photovoltaic equipment with low failure frequency; The reset operation and maintenance cycle interval duration is obtained, and the operation and maintenance cycle interval duration is updated and applied to the first operation and maintenance cycle, and the adjusted operation and maintenance plan becomes the guiding basis for the second operation and maintenance cycle, thereby achieving continuous optimization of operation and maintenance management.

7. A distributed photovoltaic power station intelligent management system based on big data monitoring, used to implement the management method according to any one of claims 1 to 6, It is characterized in that include: A data acquisition module, which is used to obtain basic information, historical operating status data, and historical operation and maintenance data of each photovoltaic device, classify and sort them, and then transmit them to the big data platform module; The big data platform module is used to receive the write request from the data acquisition module and store the classified data into the distributed file database; An operation and maintenance plan execution module, which formulates a detailed operation and maintenance plan based on the historical operation and maintenance data of the big data platform module, monitors the real-time operation status data of the photovoltaic equipment in real time during the execution of the operation and maintenance plan, and determines whether the collected real-time operation status data deviates from the normal operation status mode. If the operation status mode does not deviate from the normal, the data is uploaded to the data acquisition module; If the operating status mode deviates from normal, the suspected fault type of the photovoltaic equipment is determined, and a detailed inspection plan is formulated based on the suspected fault type. The data obtained in the inspection plan is uploaded to the data acquisition module, and the dynamic adjustment of the operation and maintenance plan for the next cycle is achieved based on the big data platform module.

8. According to claim 7, the distributed photovoltaic power station intelligent management system based on big data monitoring, It is characterized in that The data acquisition module specifically includes: A data acquisition unit, the data acquisition unit is used to acquire basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device; A data classification processing unit, wherein the data classification processing unit is used to classify and process the historical operating status data and historical operation and maintenance data of each photovoltaic device according to the basic information of each photovoltaic device; A data transmission unit, wherein the data transmission unit is used to send a write request to the big data module and transmit the classified and sorted data to the big data platform module.

9. According to claim 8, the distributed photovoltaic power station intelligent management system based on big data monitoring, It is characterized in that The big data platform module specifically includes: A data receiving unit, the data receiving unit is used to receive a write request from the data transmission unit; The data storage unit is used to store the basic information, historical operating status data and historical operation and maintenance data of each photovoltaic device after classification into a distributed file database, providing a basis for data analysis and application of the operation and maintenance plan execution module.

10. The distributed photovoltaic power station intelligent management system based on big data monitoring according to claim 9, It is characterized in that The operation and maintenance plan execution module specifically includes: An operation and maintenance plan formulation unit, the operation and maintenance plan formulation unit is used to formulate an operation and maintenance plan based on historical operation and maintenance data of the big data platform module; A real-time monitoring unit, wherein the real-time monitoring unit is used to monitor the real-time operating status data of the photovoltaic device in real time; An operation status definition unit, which is used to construct an operation status definition model for various photovoltaic devices based on the historical operation and maintenance data of the big data platform module, and to determine whether the collected real-time operation status data deviates from a normal operation status mode based on the operation status definition model by real-time monitoring of the operation status data of the photovoltaic devices; An operation data uploading unit, wherein the operation data uploading unit is used to upload the real-time operation status data of the photovoltaic device detected by the operation status definition unit and in the normal operation state mode monitored by the real-time monitoring unit to the historical operation status data of the data acquisition unit; A fault type assessment unit, which is used to obtain historical operation and maintenance data and build a fault type assessment model for various types of photovoltaic equipment, assess the fault type of the photovoltaic equipment that has been detected by the operation state definition unit and deviates from the normal operation state mode, and record it as a suspected fault type; An inspection plan formulation unit, the inspection plan formulation unit is used to formulate a detailed inspection plan according to the suspected fault type obtained by the fault type assessment unit, the inspection plan at least including: inspection time, inspection location, inspection items, required tools and materials; A fault data uploading unit, the fault data uploading unit is used to upload the fault type information and operation fault data of the photovoltaic equipment determined in the inspection plan to the historical operation and maintenance data in the data acquisition unit; An operation and maintenance cycle setting unit, which is used to set the initial operation and maintenance cycle interval according to the basic information of the photovoltaic equipment, and dynamically adjust the operation and maintenance cycle interval in combination with historical operation and maintenance data; The data updating unit is used to upload the data collected in the current operation and maintenance plan to the data acquisition unit, so that the operation status definition unit, the fault type assessment unit, and the operation and maintenance plan formulation unit can adapt to the actual conditions of the photovoltaic equipment and make adjustments.

Images