Integrated photovoltaic power station multi-dimensional operation and maintenance management method and platform

Abstract

The invention provides an integrated photovoltaic power station multi-dimensional operation and maintenance management method and platform, and relates to the technical field of photovoltaic power station management, and the method comprises the steps: obtaining a power station operation mode; traversing and starting the photovoltaic power station equipment position number list to carry out fault sample frequent pattern mining, and obtaining a photovoltaic power station equipment fault probability list; traversing and starting a photovoltaic power station equipment position number list to perform fault prediction configuration; setting a photovoltaic power station equipment operation and maintenance period list; and performing photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance period list and the fault prediction model list. The technical problem that in the prior art, due to the fact that a unified standard operation and maintenance mode is usually adopted in a photovoltaic power station, differentiated inspection configuration is not carried out according to the difference of photovoltaic equipment, and operation and maintenance management precision is low can be solved, personalized management is carried out on different equipment through differentiated inspection configuration, and operation and maintenance efficiency is improved. Missing inspection and excessive inspection are avoided, and the precision and efficiency of operation and maintenance management are improved.

Description

Technical Field

[0001] The present application relates to the technical field of photovoltaic power station management, and in particular to a multi-dimensional operation and maintenance management method and platform for an integrated photovoltaic power station. Background Art

[0002] Photovoltaic power stations are energy production facilities that use solar energy to convert into electrical energy through photovoltaic technology. The operation and maintenance management of photovoltaic power stations includes daily monitoring, inspection, fault diagnosis, maintenance, cleaning, performance optimization and other aspects of the equipment to ensure that the equipment of the power station can operate efficiently, stably and safely, and maximize the power generation efficiency and equipment life. The existing operation and maintenance management method of photovoltaic power stations uses fixed time intervals to inspect the equipment. This inspection method does not take into account the different characteristics and failure probability of each device, and all equipment is inspected at the same frequency. The unified management of photovoltaic power stations often ignores the differences in equipment, resulting in equipment with high failure risks not being repaired in time, while low-risk equipment wastes resources due to excessive maintenance, resulting in low accuracy and low efficiency of operation and maintenance management.

[0003] In summary, the prior art has a technical problem in that photovoltaic power stations usually adopt a unified standard operation and maintenance method, and do not perform differentiated inspection configurations according to differences in photovoltaic equipment, resulting in low operation and maintenance management accuracy. Summary of the invention

[0004] The purpose of this application is to provide a multi-dimensional operation and maintenance management method and platform for an integrated photovoltaic power station, so as to solve the technical problem in the prior art that photovoltaic power stations usually adopt a unified standard operation and maintenance method without differentiated inspection configuration according to the differences in photovoltaic equipment, resulting in low operation and maintenance management accuracy.

[0005] In view of the above problems, the present application provides a multi-dimensional operation and maintenance management method and platform for an integrated photovoltaic power station.

[0006] In the first aspect, the present application provides a multi-dimensional operation and maintenance management method for an integrated photovoltaic power station, which is implemented through a multi-dimensional operation and maintenance management platform for an integrated photovoltaic power station, wherein the multi-dimensional operation and maintenance management method for an integrated photovoltaic power station includes: obtaining a power station operation mode, wherein the power station operation mode has a starting photovoltaic power station equipment position number list; based on the power station operation mode, traversing the starting photovoltaic power station equipment position number list to perform frequent pattern mining of fault samples, and obtaining a photovoltaic power station equipment failure probability list; traversing the starting photovoltaic power station equipment position number list to perform fault prediction configuration, and obtaining a fault prediction model list; setting a photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; and performing photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

[0007] In the second aspect, the present application also provides an integrated photovoltaic power station multi-dimensional operation and maintenance management platform, which is used to execute an integrated photovoltaic power station multi-dimensional operation and maintenance management method as described in the first aspect, wherein the integrated photovoltaic power station multi-dimensional operation and maintenance management platform includes: an operation mode acquisition module, the operation mode acquisition module is used to obtain the power station operation mode, wherein the power station operation mode has a starting photovoltaic power station equipment position number list; a fault mining module, the fault mining module is used to traverse the starting photovoltaic power station equipment position number list based on the power station operation mode to perform frequent pattern mining of fault samples, and obtain a photovoltaic power station equipment failure probability list; a fault prediction module, the fault prediction module is used to traverse the starting photovoltaic power station equipment position number list to perform fault prediction configuration, and obtain a fault prediction model list; an operation and maintenance cycle determination module, the operation and maintenance cycle determination module is used to set the photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; an operation and maintenance management module, the operation and maintenance management module is used to perform photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: By obtaining the power station operation mode, wherein the power station operation mode has a starting photovoltaic power station equipment bit number list; based on the power station operation mode, traverse the starting photovoltaic power station equipment bit number list to perform frequent pattern mining of fault samples, and obtain a photovoltaic power station equipment failure probability list; traverse the starting photovoltaic power station equipment bit number list to perform fault prediction configuration, and obtain a fault prediction model list; set the photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; and perform photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list. In other words, by performing frequent pattern mining on the fault data of the power station equipment, it is identified which equipment is more likely to fail, as well as the high-frequency mode of equipment failure, and according to the historical data, operating environment, and failure mode of each equipment, the failure probability of the equipment is predicted, the operation and maintenance cycle of each equipment is dynamically adjusted, and the equipment operation and maintenance plan is automatically generated, thereby improving the accuracy and efficiency of operation and maintenance management.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 A schematic diagram of a multi-dimensional operation and maintenance management method for an integrated photovoltaic power station is provided for this application; Figure 2 The present invention is a structural schematic diagram of a multi-dimensional operation and maintenance management platform for an integrated photovoltaic power station.

[0012] Explanation of the reference numerals: operating mode acquisition module 11 , fault mining module 12 , fault prediction module 13 , operation and maintenance cycle determination module 14 , operation and maintenance management module 15 . DETAILED DESCRIPTION

[0013] This application provides a multi-dimensional operation and maintenance management method and platform for an integrated photovoltaic power station, which solves the technical problem in the prior art that photovoltaic power stations usually adopt a unified standard operation and maintenance method, and do not perform differentiated inspection configurations according to the differences in photovoltaic equipment, resulting in low operation and maintenance management accuracy. By mining the frequent pattern of the fault data of power station equipment, we can identify which equipment is more prone to failure and the high-frequency patterns of equipment failures. Based on the historical data, operating environment, and failure mode of each device, we can predict the failure probability of the device, dynamically adjust the operation and maintenance cycle of each device, and automatically generate the operation and maintenance plan of the equipment, thereby improving the accuracy and efficiency of operation and maintenance management.

[0014] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] For example, please refer to the attached Figure 1 The present application provides a multi-dimensional operation and maintenance management method for an integrated photovoltaic power station, wherein the multi-dimensional operation and maintenance management method for an integrated photovoltaic power station is applied to a multi-dimensional operation and maintenance management platform for an integrated photovoltaic power station, and the multi-dimensional operation and maintenance management method for an integrated photovoltaic power station specifically includes the following steps: Step 1: Obtain the power plant operation mode, wherein the power plant operation mode has a starting photovoltaic power plant equipment bit number list.

[0016] Specifically, the power station operation mode refers to the behavior mode or manifestation of the power station equipment under different operating conditions. Different operation modes are determined by analyzing the operation data of the power station equipment. The operation data of all equipment in the photovoltaic power station are collected and analyzed, and different operation modes are defined and classified, including the startup state, operation state, fault state, etc. of the equipment. According to the operation mode of the power station, identify which equipment has been started in this mode. The started equipment will be recorded and extracted according to its unique equipment bit number to generate a bit number list of the started equipment. Starting the photovoltaic power station equipment refers to the power-on or activation state of the equipment in the power station, such as photovoltaic modules, inverters, monitoring systems, etc. Starting the equipment usually means that the equipment starts working and starts to produce electricity. The equipment bit number (or equipment number) refers to the unique identifier of each equipment in the power station. Each equipment has a unique number to distinguish different equipment in the power station. The equipment bit number list is a list of all the bit numbers of the started photovoltaic power station equipment. Through the power station operation mode, the bit numbers of all started equipment are automatically identified and recorded, which helps to reduce errors and omissions in management and improve the accuracy of equipment management.

[0017] Step 2: Based on the power station operation mode, traverse the starting photovoltaic power station equipment bit number list to perform frequent pattern mining of fault samples to obtain a photovoltaic power station equipment failure probability list.

[0018] Specifically, according to the operating mode of the power station, the list of all the photovoltaic equipment numbers that are started in the photovoltaic power station (that is, all the running equipment) is traversed. The fault data of each device is collected, including the type of fault, the time of occurrence, the duration of the fault, and other information. For each device, all fault samples that occur in the device within a certain period of time are collected, including the time when the fault occurs, the operating status of the device (such as temperature, voltage, etc.), and the type of fault (such as inverter failure, photovoltaic module failure, etc.), the frequency of the fault, etc. The frequent item set mining algorithm is used to analyze the fault samples of the equipment to identify which fault modes frequently occur on multiple devices. Fault sample frequent pattern mining is a technology in data mining, which aims to find frequently repeated fault modes or fault patterns in a large amount of equipment failure data.

[0019] According to the results of frequent pattern mining, statistics are used to calculate the failure probability of each device. The failure frequencies of each device are summed to obtain the sum of the failure frequencies. The failure probability of each device is obtained by calculating the ratio of the failure frequency of each device to the sum of the failure frequencies. The failure probabilities of all devices are sorted into a failure probability list, and the failure risk of each device is displayed based on the failure history of the device and the frequency of the failure mode. Through frequent pattern mining, common failure modes in multiple devices are identified, the historical failure data and frequent failure modes of the devices are analyzed, the failure probability of each device is calculated, and it is determined which devices need to be repaired or maintained first.

[0020] Step 3: Traverse the starting photovoltaic power station equipment number list to perform fault prediction configuration and obtain a fault prediction model list.

[0021] Specifically, each photovoltaic power station device in the starting photovoltaic power station device number list is configured for fault prediction, and a fault prediction model is configured for each photovoltaic power station device, and finally a fault prediction model list is obtained, which contains a collection of multiple device fault prediction models, and each model corresponds to one device. First, the service time and maintenance times corresponding to each device are obtained from the starting photovoltaic power station device number list. The service time refers to the cumulative operating time of the photovoltaic power station equipment from the time it is put into use to the current moment, usually in hours or days, reflecting the degree of use and operating status of the equipment. The number of maintenance times refers to the number of maintenance times that occur to photovoltaic power station equipment during its service life, which usually represents the failure frequency of the equipment or the number of times maintenance is required, and is an important indicator of equipment reliability.

[0022] The historical predicted fault record data and historical actual fault record data of each device that meet the service time and maintenance times of each device are obtained from the fault prediction record log, and the fault prediction accuracy of each device is calculated. The fault prediction record log is the log data that records the historical fault occurrence of the device, usually including the operating status of the device, the time when the fault occurred, the type of fault, etc. According to the fault prediction accuracy of each device, the recommended number of fault prediction unit integration is configured. More integrated units help improve the accuracy and stability of the fault prediction model, but also increase the consumption of computing resources. The number of integrated units is automatically adjusted according to the actual prediction performance of the device, the intelligent level of operation and maintenance management is improved, and equipment failures and downtime are reduced. Fault prediction configuration refers to configuring a fault prediction model for each photovoltaic power station device, including selecting a suitable prediction method, setting parameters, selecting a training data set, etc., and finally generating a model that can predict equipment failures.

[0023] According to the recommended number of fault prediction unit integrations, configure the fault prediction unit of each device, including adjusting the parameters of the model (such as learning rate, regularization parameters, etc.) or selecting different algorithms for adjustment, and save the fault prediction model of each device to ensure that each device has a specific prediction model. After completing the configuration of the fault prediction model of each device, add it to the fault prediction model list, which includes the fault prediction models of all photovoltaic power station equipment. By configuring the fault prediction model, the photovoltaic power station can warn of possible faults in advance, reduce unexpected downtime, and improve operation and maintenance efficiency. Each device has its own fault prediction model, which can make personalized predictions based on the specific conditions of the device (such as service life, maintenance records, etc.), improving the accuracy of the prediction.

[0024] Step 4: Setting a photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list.

[0025] Specifically, the photovoltaic power station equipment failure probability list contains a list of the failure probabilities of all equipment in the power station. The failure probability of each device indicates the possibility of failure of the device within a certain period of time. The photovoltaic power station equipment operation and maintenance cycle list is an operation and maintenance cycle list corresponding to the failure probability list. The operation and maintenance cycle of each device is set based on its failure probability, with the aim of ensuring that high-risk equipment is frequently inspected and maintained, while the maintenance of low-risk equipment can be relatively reduced. The operation and maintenance cycle can include real-time monitoring, scheduled inspections, etc. According to the failure probability of each device, an appropriate operation and maintenance cycle is automatically set for each device based on preset rules, and different inspection frequencies correspond to different failure probabilities.

[0026] The operation and maintenance cycle of each device is set according to the failure probability through a set of preset rules. The failure probability is divided into several intervals, and the corresponding operation and maintenance cycle is given according to different intervals. The lower the failure probability, the shorter the operation and maintenance cycle and the higher the inspection frequency. The preset rules are pre-set rules used to determine the operation and maintenance cycle that should correspond to each failure probability. Finally, a list of operation and maintenance cycles for photovoltaic power station equipment is formed, and each device has its corresponding operation and maintenance cycle. According to the different failure probabilities of the equipment, the inspection frequency of each device is dynamically adjusted to avoid unnecessary frequent inspections, while also ensuring that high-risk equipment is handled in a timely manner.

[0027] Step 5: Perform photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

[0028] Specifically, the operation and maintenance cycle of each photovoltaic power station equipment is determined by the failure probability, which determines the frequency of inspection, maintenance or monitoring of each equipment. The cycle includes real-time monitoring, regular inspection (such as every 10 minutes, 30 minutes, 24 hours, etc.) and longer inspection cycles (such as 7 days). The fault prediction model list includes the fault prediction model corresponding to each device. According to the fault prediction model and operation and maintenance cycle list of each device, a specific operation and maintenance plan is set for each device, which depends on the predicted failure probability of the device and the status of the device, aiming to minimize the occurrence of failures and improve operation and maintenance efficiency. According to the fault prediction model of each device, the failure risk of the device is dynamically evaluated. For devices with a higher probability of failure, more monitoring and inspection are arranged. For devices with a lower probability of failure, the inspection cycle can be extended, for example, once every 24 hours or every 7 days.

[0029] The fault prediction model will continuously track and evaluate the health status of the equipment. When the model predicts that the equipment has a high risk of failure, it will issue an early warning, adjust the equipment's operation and maintenance cycle, and increase the inspection frequency to ensure timely processing before the equipment fails. The equipment's operation and maintenance plan is dynamically adjusted based on the latest data and operation and maintenance cycle of the equipment failure prediction. In the actual operation and maintenance process, the data of each inspection and maintenance is collected and fed back to the fault prediction model for further optimization, and the model is continuously improved to improve the accuracy and reliability of fault prediction, thereby providing a more accurate decision-making basis for future operation and maintenance management. By combining the fault prediction model and the operation and maintenance cycle, a customized operation and maintenance management plan is provided for each device to ensure that high-risk equipment receives full attention, reduce the occurrence rate of equipment failures, effectively avoid excessive inspections or delayed inspections, thereby reducing the workload, and ensuring that the equipment is inspected and maintained in a timely manner.

[0030] Further, step 2 of this application includes: According to the starting photovoltaic power station equipment bit number list, extract the photovoltaic power station equipment with the first bit number to the photovoltaic power station equipment with the Nth bit number; collect the first historical fault frequency in the preset time zone when the photovoltaic power station equipment with the first bit number is in the power station operation mode; until collecting the Nth historical fault frequency in the preset time zone when the photovoltaic power station equipment with the Nth bit number is in the power station operation mode; count the sum of the frequencies of the first historical fault frequency in the preset time zone to the Nth historical fault frequency in the preset time zone to obtain the sum of fault frequencies; traverse the first historical fault frequency in the preset time zone to the Nth historical fault frequency in the preset time zone, calculate the ratio with the sum of fault frequencies respectively, set it as the failure probability of the photovoltaic power station equipment with the first bit number to the failure probability of the photovoltaic power station equipment with the Nth bit number, and add it into the photovoltaic power station equipment failure probability list.

[0031] Specifically, extract the first photovoltaic power station device to the Nth photovoltaic power station device from the starting photovoltaic power station device bit number list, that is, all the devices in the list. N is an integer representing the total number of devices. The preset time zone refers to a time period pre-set according to the operation requirements of the power station, during which the historical fault frequency data of the equipment is collected. The preset time zone can be a specific time range such as a day, a week, a month, etc., and the specific duration is set by the power station operation and maintenance personnel according to the needs. Obtain the historical fault frequency of the first photovoltaic power station device in the preset time zone, that is, the frequency of failure of the first photovoltaic power station device in the preset time zone. The fault frequency refers to the number of times the equipment fails in the preset time zone, usually expressed as the ratio of the number of fault events to the length of the time period.

[0032] Similarly, the fault frequencies of all photovoltaic power station equipment in the preset time zone in the power station operation mode are collected until the Nth historical fault frequency in the preset time zone is obtained. The historical fault frequencies of the first-numbered device to the Nth-numbered device in the preset time zone are summed to obtain the sum of the fault frequencies. The first historical fault frequency in the preset time zone is traversed until the Nth historical fault frequency in the preset time zone, and the ratios are calculated with the sum of the fault frequencies respectively, the failure probability of each device is calculated, and the historical fault frequency of each device is compared with the sum of the fault frequencies of all devices to obtain the failure probability of the first-numbered photovoltaic power station equipment until the Nth-numbered photovoltaic power station equipment. For example, there are three devices (N=3), whose historical fault frequencies are 0.02, 0.05, and 0.01 respectively, then the sum of the fault frequencies is 0.08, then the failure probability of device 1 is 0.25, the failure probability of device 2 is 0.625, and the failure probability of device 3 is 0.125.

[0033] Calculate the failure probability of each device, and add the failure probability of the first photovoltaic power station device to the Nth photovoltaic power station device failure probability list. The photovoltaic power station device failure probability list contains the failure probability of each device in the power station, which is used to assess the failure risk of each device and determine whether the device needs more frequent maintenance. By calculating the ratio of the historical failure frequency to the sum of the device failure frequencies, the failure probability of each device is accurately assessed, and the high failure probability devices are given priority for key inspection and maintenance, reducing the risk of equipment failure and improving the efficiency of operation and maintenance management.

[0034] Further, step three of this application includes: According to the list of starting photovoltaic power station equipment numbers, extract the photovoltaic power station equipment with the first number; obtain the first service time and the first maintenance times of the photovoltaic power station equipment with the first number; use the first service time, the first maintenance times and the model of the photovoltaic power station equipment with the first number as the equipment portrait, perform prediction accuracy analysis based on the fault prediction record log, and obtain the fault prediction accuracy of the photovoltaic power station with the first number; configure the integrated number of fault prediction units according to the fault prediction accuracy of the photovoltaic power station with the first number, and obtain the recommended integrated number of fault prediction units; configure the integrated model of the fault prediction unit according to the recommended integrated number of fault prediction units, generate a fault prediction model for the photovoltaic power station equipment with the first number, and add it to the fault prediction model list.

[0035] Specifically, a photovoltaic power station device is randomly selected from the starting photovoltaic power station device position number list as the first photovoltaic power station device. The first service time and the first maintenance number of the first photovoltaic power station device are obtained. The service time refers to the cumulative operating time of the photovoltaic power station device from the time it is put into use to the current moment, usually in hours or days, which can reflect the use degree and operating status of the equipment. The first maintenance number refers to the number of maintenance times that the first photovoltaic power station device has undergone during its service, which usually represents the failure frequency of the equipment or the number of times maintenance is required, and is an important indicator of equipment reliability.

[0036] Through the first service time, the first maintenance number and the first photovoltaic power station equipment model, a portrait is established for the photovoltaic power station equipment, which integrates the historical use of the equipment, maintenance records and model information to provide background data for the fault prediction model. The historical predicted fault record data and historical actual fault record data of the first photovoltaic power station equipment model that meet the first service time and the first maintenance number in the fault prediction record log are counted. The number of intersection fault types and the number of union fault types of the historical predicted fault record data and the historical actual fault record data of the first photovoltaic power station equipment model are counted, that is, the correctly predicted fault tree and the total fault tree, and the fault prediction accuracy of the first photovoltaic power station is calculated based on this.

[0037] According to the calculated first-number PV power station equipment fault prediction accuracy, configure the recommended number of fault prediction unit integrations. When the prediction accuracy is high, reduce the number of integrated units; when the prediction accuracy is low, more prediction units need to be added to improve the reliability of fault prediction. If the prediction accuracy of the equipment is very high, configure fewer prediction units, and 2 or 3 prediction units may be recommended. If the prediction accuracy is low, more prediction units are needed to enhance the prediction effect, and 5 or more prediction units may be recommended.

[0038] According to the recommended number of fault prediction unit integrations, configure the integrated model of the fault prediction unit by adjusting the complexity of the neural network (such as the number of network layers, the number of nodes in each layer, the activation function), adding model instances, etc. The prediction accuracy can be improved by adding model instances or increasing the complexity of the neural network, but the ultimate goal is usually to integrate these different configurations into a single prediction model, that is, the number one photovoltaic power station equipment fault prediction model. Add the generated number one photovoltaic power station equipment fault prediction model to the fault prediction model list. The fault prediction model list includes fault prediction models for multiple devices, each of which is a personalized model for different devices. Personalized predictions are made for the service life and maintenance records of each device to provide more accurate operation and maintenance suggestions, ensuring that the fault prediction model of each photovoltaic power station equipment is optimized based on its specific operation history and performance characteristics, thereby improving the efficiency and effectiveness of overall operation and maintenance management.

[0039] Furthermore, the present application also includes the following steps: The historical predicted fault record data of the first-number photovoltaic power station equipment model and the historical actual fault record data of the first-number photovoltaic power station equipment model of the fault prediction record log that meet the first service time and the first maintenance number are counted; the ratio of the number of intersection fault types of the first-number photovoltaic power station equipment model historical predicted fault record data and the first-number photovoltaic power station equipment model historical actual fault record data to the number of union fault types is counted, and is set as the fault prediction accuracy rate of the first-number photovoltaic power station.

[0040] Specifically, the fault prediction record log is log data used to track the history of equipment fault predictions, and contains the prediction results generated by the fault prediction model, including the predicted fault type, prediction time, prediction accuracy, etc. All historical predicted fault data related to the first-numbered device model are extracted from the fault prediction record log, including the predicted fault type and prediction time. Similarly, all actual fault data related to the first-numbered device model are extracted from the fault prediction record log, including the actual fault type and occurrence time.

[0041] The number of fault types in the intersection of the historical predicted fault record data of the first photovoltaic power station equipment model and the historical actual fault record data of the first photovoltaic power station equipment model is counted, that is, the number of fault types that have appeared in both the historical fault prediction and the historical actual fault, which is the overlap between the fault type predicted by the model and the actual fault type. The number of fault types in the union of the historical predicted fault record data of the first photovoltaic power station equipment model and the historical actual fault record data of the first photovoltaic power station equipment model is counted, that is, the total number of all fault types in the historical predicted fault record and the historical actual fault record, including the intersection part and the non-intersection part, that is, including the same type and different types.

[0042] By calculating the ratio of the number of intersection fault types to the number of union fault types, we can get the fault prediction accuracy, which is an important indicator for evaluating the performance of the fault prediction model. It refers to the overlap between the fault types predicted by the model and the actual fault types. By calculating the fault prediction accuracy, we can quantify the effect of the fault prediction model and evaluate the performance of the current fault prediction model.

[0043] Furthermore, the present application also includes the following steps: Calculate the ratio of the accuracy threshold to the fault prediction accuracy of the first-numbered photovoltaic power station to obtain a first integration factor; round down the first integration factor to obtain a first integration index; use the first integration index as an exponent of 2 to calculate the recommended fault prediction unit integration quantity.

[0044] Specifically, setting an accuracy threshold based on past experience, industry standards, or expected maintenance needs is a preset standard used to measure the minimum acceptable accuracy of the fault prediction model. The ratio of the accuracy threshold to the first-number PV power station fault prediction accuracy is calculated to obtain the first integration factor, which reflects the difference between the actual prediction accuracy and the expected accuracy threshold. When the first-number PV power station fault prediction accuracy is high, it means that the model has a strong prediction capability, so the accuracy of the fault prediction is higher. The larger the denominator, the smaller the first integration factor should be. The first integration factor is rounded down to obtain the first integration index, which means that the decimal part of the first integration factor is discarded and the integer part is retained to ensure reasonable resource allocation during the integration process and avoid over-configuration of integrated units.

[0045] The first integration factor is a ratio, usually a decimal, and its meaning is to measure whether the prediction model has achieved the expected accuracy. Since the role of the integration factor is to determine the number of integrated units, the decimal part cannot be used to directly configure resources. The rounding operation can convert the configuration amount of model complexity into a more concise integer, ensuring that the number of integrated units is appropriate and resources are not wasted. If the first integration factor is large, rounding makes the integrated model configuration within a reasonable range, avoiding the waste of computing resources due to excessive complexity of the model. If the first integration factor is small, the first integration index is 0 after adjustment, indicating that the prediction accuracy of the device is already very high, and it is recommended not to add additional integrated units.

[0046] The first integration index is used as the exponent of 2 to calculate the recommended number of fault prediction unit integrations. The exponent of 2 is a very effective way to achieve exponential growth, which can ensure that the complexity and prediction accuracy of the model can also be moderately improved as the integration index increases. In ensemble learning, the number of integration units (such as decision trees, neural networks, etc.) generally grows exponentially, especially in models such as reinforcement learning and random forests. Using an exponent of 2 means that as the number of integration units increases, the complexity and prediction performance of the model will increase exponentially. This growth method makes the configuration very flexible and adjustable. For each additional integration unit (that is, the first integration index increases by 1), the number of integration units doubles. Through this exponential growth method, resources can be reasonably allocated, and the number of integration units can be flexibly configured according to the prediction accuracy requirements of different devices. If the fault prediction accuracy of the device is high, the recommended number of integration units is small; otherwise, the number of integration units is increased to improve the prediction accuracy.

[0047] By rounding down, the decimal part in the first integration factor is eliminated, ensuring that the number of integrated units will not be too refined and preventing waste of computing resources. Through exponential calculation of 2, the model can flexibly adjust the number of integrated units according to the change of the first integration index. The exponential growth allows the model configuration to adapt to different accuracy requirements, providing a better balance between performance and resources. The combination of rounding down and exponential calculation ensures the adaptability of the fault prediction model. The complexity of the model will not be too simple due to low prediction accuracy, nor will it waste resources due to high accuracy. The operation and maintenance management system of the photovoltaic power station can dynamically optimize its fault prediction model, improve the accuracy of equipment fault prediction, and make rational use of computing resources.

[0048] Furthermore, the present application also includes the following steps: Collect historical operating status data and historical fault probability identification data of the first-number photovoltaic power station equipment model; use the historical fault probability identification data as supervision and the historical operating status data as input to configure a BP neural network to generate the fault prediction unit.

[0049] Specifically, collect the historical operating status data of the first-number photovoltaic power station equipment model, that is, the operating data in the past period of time, including the operating parameters of the equipment, such as voltage, current, power, temperature, humidity, etc. In order to ensure the accuracy of the prediction, it should be as detailed as possible and have a long time span. Collect the historical fault probability identification data of the first-number photovoltaic power station equipment model. The data tags related to the historical faults of the equipment are usually used to identify whether the equipment has failed. It is usually in binary form, indicating whether the equipment has failed at a certain point in time (1 indicates failure, 0 indicates no failure).

[0050] Using historical operating status data as input features (feature vectors) and historical fault probability identification data as target output, a BP neural network is configured for supervised learning. BP neural networks are widely used in classification and regression problems. Through multi-layer neuron connections, input data is passed to the output layer and errors are calculated, and then the weights in the network are updated through the back propagation algorithm. BP neural networks can handle complex nonlinear relationships and are suitable for complex equipment fault prediction tasks such as photovoltaic power stations.

[0051] Neural networks use input operating status data to predict whether a device will fail. Determine the network structure, including the number of input layer nodes (corresponding to the number of features of the operating status data), the number of hidden layer nodes, and the number of output layer nodes (usually 1, indicating the probability of failure). The input layer is used to receive various operating status data of the device, such as voltage, current, temperature, etc.; the hidden layer uses multiple neurons to perform nonlinear conversion to explore the potential relationship between the operating status of the device and the occurrence of a fault; the output layer outputs the probability of a fault, usually a value, indicating the possibility of a fault.

[0052] The network weights and deviations are continuously adjusted based on historical operating status data and historical fault probability identification data. Through the back propagation algorithm, the network parameters are updated based on the prediction error, making the predicted value closer to the actual value, thereby improving the prediction accuracy. The trained BP neural network model forms a fault prediction unit, which inputs the current operating status data of the equipment and predicts the probability of equipment failure. Through the generated fault prediction unit, the photovoltaic power station can monitor and predict equipment failures in real time during equipment operation, improving the equipment's operation and maintenance efficiency and management level.

[0053] Further, step 4 of this application includes: When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.5, the operation and maintenance cycle of the photovoltaic power station equipment is set to real-time monitoring; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.3 and less than 0.5, the operation and maintenance cycle of the photovoltaic power station equipment is set to 10 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.2 and less than 0.3, the operation and maintenance cycle of the photovoltaic power station equipment is set to 30 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.1 and less than 0.2, the operation and maintenance cycle of the photovoltaic power station equipment is set to 60 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.05 and less than 0.1, the operation and maintenance cycle of the photovoltaic power station equipment is set to 24 hours; when the failure probability of the photovoltaic power station equipment is less than 0.05, the operation and maintenance cycle of the photovoltaic power station equipment is set to 7 days.

[0054] Specifically, the failure probability of photovoltaic power station equipment is obtained by collecting the historical failure frequency of each photovoltaic power station equipment number list (i.e., equipment identifier), performing weighted calculation, and calculating the ratio with the failure frequency of the overall equipment, and finally obtaining the failure probability of each device. The photovoltaic power station equipment operation and maintenance cycle list assigns an operation and maintenance cycle to each device, indicating the inspection, maintenance or monitoring frequency of the device. The length of the operation and maintenance cycle is determined by the failure probability of the equipment. The higher the failure probability, the shorter the operation and maintenance cycle, and vice versa.

[0055] When the probability of PV power station equipment failure is greater than or equal to 0.5, the PV power station equipment operation and maintenance cycle is set to real-time monitoring. The risk of equipment failure is very high, and the equipment status must be continuously monitored to be ready for intervention and maintenance at any time; when the probability of PV power station equipment failure is greater than or equal to 0.3 and less than 0.5, the PV power station equipment operation and maintenance cycle is set to 10 minutes, indicating that the probability of equipment failure is high. Although real-time monitoring is not required, frequent inspections are still required to prevent equipment failure; when the probability of PV power station equipment failure is greater than or equal to 0.2 and less than 0.3, the PV power station equipment operation and maintenance cycle is set to 30 minutes, indicating that the equipment The failure risk is medium, and moderately frequent inspection and monitoring are required; when the failure probability of PV power station equipment is greater than or equal to 0.1 and less than 0.2, the operation and maintenance cycle of PV power station equipment is set to 60 minutes, indicating that the failure probability of the equipment is low and the inspection frequency can be reduced; when the failure probability of PV power station equipment is greater than or equal to 0.05 and less than 0.1, the operation and maintenance cycle of PV power station equipment is set to 24 hours, indicating that the equipment is in a relatively low failure risk state; when the failure probability of PV power station equipment is less than 0.05, the operation and maintenance cycle of PV power station equipment is set to 7 days, indicating that the failure probability of the equipment is very low and frequent inspection is not required.

[0056] Dynamic operation and maintenance cycles are set according to the failure probability of the equipment to achieve more efficient resource scheduling. For equipment with a high probability of failure, the operation and maintenance frequency is automatically increased to ensure that the failure can be discovered and handled as early as possible, and the inspection frequency of equipment with a low probability of failure is automatically reduced, thereby reducing unnecessary inspection and maintenance costs, while ensuring timely handling of high-risk equipment. As the probability of equipment failure changes, the operation and maintenance strategy is flexibly adjusted to ensure that the operation and maintenance plan of the entire power station is always consistent with the actual situation of the equipment.

[0057] In summary, the multi-dimensional operation and maintenance management method of an integrated photovoltaic power station provided by this application has the following technical effects: By obtaining the power station operation mode, wherein the power station operation mode has a starting photovoltaic power station equipment bit number list; based on the power station operation mode, traverse the starting photovoltaic power station equipment bit number list to perform frequent pattern mining of fault samples, and obtain a photovoltaic power station equipment failure probability list; traverse the starting photovoltaic power station equipment bit number list to perform fault prediction configuration, and obtain a fault prediction model list; set the photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; and perform photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list. In other words, by performing frequent pattern mining on the fault data of the power station equipment, it is identified which equipment is more likely to fail, as well as the high-frequency mode of equipment failure, and according to the historical data, operating environment, and failure mode of each equipment, the failure probability of the equipment is predicted, the operation and maintenance cycle of each equipment is dynamically adjusted, and the equipment operation and maintenance plan is automatically generated, thereby improving the accuracy and efficiency of operation and maintenance management.

[0058] Embodiment 2, based on the same inventive concept as the integrated photovoltaic power station multi-dimensional operation and maintenance management method in the aforementioned embodiment 1, the present application also provides an integrated photovoltaic power station multi-dimensional operation and maintenance management platform, please refer to the attached Figure 2 The integrated photovoltaic power station multi-dimensional operation and maintenance management platform includes: An operation mode acquisition module 11, the operation mode acquisition module 11 is used to obtain the operation mode of the power station, wherein the operation mode of the power station has a starting photovoltaic power station equipment bit number list; a fault mining module 12, the fault mining module 12 is used to traverse the starting photovoltaic power station equipment bit number list based on the power station operation mode to perform frequent pattern mining of fault samples, and obtain a photovoltaic power station equipment failure probability list; a fault prediction module 13, the fault prediction module 13 is used to traverse the starting photovoltaic power station equipment bit number list to perform fault prediction configuration, and obtain a fault prediction model list; an operation and maintenance cycle determination module 14, the operation and maintenance cycle determination module 14 is used to set the photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; an operation and maintenance management module 15, the operation and maintenance management module 15 is used to perform photovoltaic power station operation and maintenance management according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

[0059] Furthermore, the fault mining module 12 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: According to the starting photovoltaic power station equipment bit number list, extract the photovoltaic power station equipment with the first bit number to the photovoltaic power station equipment with the Nth bit number; collect the first historical fault frequency in the preset time zone when the photovoltaic power station equipment with the first bit number is in the power station operation mode; until collecting the Nth historical fault frequency in the preset time zone when the photovoltaic power station equipment with the Nth bit number is in the power station operation mode; count the sum of the frequencies of the first historical fault frequency in the preset time zone to the Nth historical fault frequency in the preset time zone to obtain the sum of fault frequencies; traverse the first historical fault frequency in the preset time zone to the Nth historical fault frequency in the preset time zone, calculate the ratio with the sum of fault frequencies respectively, set it as the failure probability of the photovoltaic power station equipment with the first bit number to the failure probability of the photovoltaic power station equipment with the Nth bit number, and add it into the photovoltaic power station equipment failure probability list.

[0060] Furthermore, the fault prediction module 13 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: According to the list of starting photovoltaic power station equipment numbers, extract the photovoltaic power station equipment with the first number; obtain the first service time and the first maintenance times of the photovoltaic power station equipment with the first number; use the first service time, the first maintenance times and the model of the photovoltaic power station equipment with the first number as the equipment portrait, perform prediction accuracy analysis based on the fault prediction record log, and obtain the fault prediction accuracy of the photovoltaic power station with the first number; configure the integrated number of fault prediction units according to the fault prediction accuracy of the photovoltaic power station with the first number, and obtain the recommended integrated number of fault prediction units; configure the integrated model of the fault prediction unit according to the recommended integrated number of fault prediction units, generate a fault prediction model for the photovoltaic power station equipment with the first number, and add it to the fault prediction model list.

[0061] Furthermore, the fault prediction module 13 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: The historical predicted fault record data of the first-number photovoltaic power station equipment model and the historical actual fault record data of the first-number photovoltaic power station equipment model of the fault prediction record log that meet the first service time and the first maintenance number are counted; the ratio of the number of intersection fault types of the first-number photovoltaic power station equipment model historical predicted fault record data and the first-number photovoltaic power station equipment model historical actual fault record data to the number of union fault types is counted, and is set as the fault prediction accuracy rate of the first-number photovoltaic power station.

[0062] Furthermore, the fault prediction module 13 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: Calculate the ratio of the accuracy threshold to the fault prediction accuracy of the first-numbered photovoltaic power station to obtain a first integration factor; round down the first integration factor to obtain a first integration index; use the first integration index as an exponent of 2 to calculate the recommended fault prediction unit integration quantity.

[0063] Furthermore, the fault prediction module 13 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: Collect historical operating status data and historical fault probability identification data of the first-number photovoltaic power station equipment model; use the historical fault probability identification data as supervision and the historical operating status data as input to configure a BP neural network to generate the fault prediction unit.

[0064] Furthermore, the operation and maintenance cycle determination module 14 in the integrated photovoltaic power station multi-dimensional operation and maintenance management platform is also used for: When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.5, the operation and maintenance cycle of the photovoltaic power station equipment is set to real-time monitoring; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.3 and less than 0.5, the operation and maintenance cycle of the photovoltaic power station equipment is set to 10 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.2 and less than 0.3, the operation and maintenance cycle of the photovoltaic power station equipment is set to 30 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.1 and less than 0.2, the operation and maintenance cycle of the photovoltaic power station equipment is set to 60 minutes; when the failure probability of the photovoltaic power station equipment is greater than or equal to 0.05 and less than 0.1, the operation and maintenance cycle of the photovoltaic power station equipment is set to 24 hours; when the failure probability of the photovoltaic power station equipment is less than 0.05, the operation and maintenance cycle of the photovoltaic power station equipment is set to 7 days.

[0065] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The integrated photovoltaic power station multi-dimensional operation and maintenance management method and specific examples in the first embodiment are also applicable to the integrated photovoltaic power station multi-dimensional operation and maintenance management platform of the present embodiment. Through the above detailed description of the integrated photovoltaic power station multi-dimensional operation and maintenance management method, those skilled in the art can clearly know the integrated photovoltaic power station multi-dimensional operation and maintenance management platform of the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the platform disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0066] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0067] 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 belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station, characterized in that: include: Obtaining a power plant operation mode, wherein the power plant operation mode has a starting photovoltaic power plant equipment bit number list; Based on the power station operation mode, traverse the starting photovoltaic power station equipment bit number list to perform frequent pattern mining of fault samples to obtain a photovoltaic power station equipment failure probability list; Traversing the starting photovoltaic power station equipment bit number list to perform fault prediction configuration and obtain a fault prediction model list; Setting a photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; The photovoltaic power station operation and maintenance management is performed according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

2. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station according to claim 1, characterized in that: Based on the power station operation mode, the starting photovoltaic power station equipment bit number list is traversed to perform frequent pattern mining of fault samples, and a photovoltaic power station equipment failure probability list is obtained, including: According to the starting photovoltaic power station equipment number list, extract the photovoltaic power station equipment with the first number to the photovoltaic power station equipment with the Nth number; Collecting a first historical fault frequency in a preset time zone when the photovoltaic power station equipment with the first digit is in the power station operation mode; Until the Nth historical fault frequency in the preset time zone when the Nth photovoltaic power station equipment is in the power station operation mode is collected; Counting the sum of the frequencies of the first historical fault frequency in the preset time zone to the Nth historical fault frequency in the preset time zone to obtain a fault frequency sum value; Traverse the first historical fault frequency in the preset time zone until the Nth historical fault frequency in the preset time zone, calculate the ratio with the sum of the fault frequencies respectively, set it as the first photovoltaic power station equipment failure probability until the Nth photovoltaic power station equipment failure probability, and add it to the photovoltaic power station equipment failure probability list.

3. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station as claimed in claim 1, characterized in that: Traverse the starting photovoltaic power station equipment number list to perform fault prediction configuration, and obtain a fault prediction model list, including: According to the starting photovoltaic power station equipment number list, extract the photovoltaic power station equipment with the first number; Obtaining the first service time and the first maintenance times of the photovoltaic power station equipment with the first digit; Taking the first service time, the first maintenance times and the first photovoltaic power station equipment model as the equipment portrait, performing prediction accuracy analysis based on the fault prediction record log to obtain the first photovoltaic power station fault prediction accuracy; According to the first-number photovoltaic power station fault prediction accuracy, configure the integrated number of fault prediction units to obtain a recommended integrated number of fault prediction units; The fault prediction unit is configured as an integrated model according to the recommended integrated number of fault prediction units, a first-number photovoltaic power station equipment fault prediction model is generated, and the model is added to the fault prediction model list.

4. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station as claimed in claim 3, characterized in that: The first service time, the first maintenance times and the first photovoltaic power station equipment model are used as equipment portraits, and prediction accuracy analysis is performed based on the fault prediction record log to obtain the first photovoltaic power station fault prediction accuracy, including: Collecting statistics on the historical predicted fault record data of the first photovoltaic power station equipment model and the historical actual fault record data of the first photovoltaic power station equipment model that meet the first service time and the first maintenance times in the fault prediction record log; The ratio of the number of intersection fault types of the historical predicted fault record data of the first-numbered photovoltaic power station equipment model and the number of union fault types of the historical actual fault record data of the first-numbered photovoltaic power station equipment model is counted, and is set as the fault prediction accuracy rate of the first-numbered photovoltaic power station.

5. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station as claimed in claim 3, characterized in that: According to the first-number photovoltaic power station fault prediction accuracy, the fault prediction unit integration quantity is configured to obtain the recommended fault prediction unit integration quantity, including: Calculate the ratio of the accuracy threshold to the first-numbered photovoltaic power station fault prediction accuracy to obtain a first integration factor; Rounding down the first integration factor to obtain a first integration index; The recommended fault prediction unit integration quantity is calculated by taking the first integration index as an exponent of 2.

6. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station as claimed in claim 3, characterized in that: The fault prediction unit construction step comprises: Collect historical operating status data and historical fault probability identification data of the first-number photovoltaic power station equipment model; The historical fault probability identification data is used as supervision, the historical operating status data is used as input, a BP neural network is configured, and the fault prediction unit is generated.

7. A multi-dimensional operation and maintenance management method for an integrated photovoltaic power station according to claim 1, characterized in that: The photovoltaic power station equipment operation and maintenance cycle list is set according to the photovoltaic power station equipment failure probability list, including: When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.5, the operation and maintenance cycle of the photovoltaic power station equipment is set to real-time monitoring; When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.3 and less than 0.5, the operation and maintenance period of the photovoltaic power station equipment is set to 10 minutes; When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.2 and less than 0.3, the operation and maintenance period of the photovoltaic power station equipment is set to 30 minutes; When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.1 and less than 0.2, the operation and maintenance period of the photovoltaic power station equipment is set to 60 minutes; When the failure probability of the photovoltaic power station equipment is greater than or equal to 0.05 and less than 0.1, the operation and maintenance cycle of the photovoltaic power station equipment is set to 24 hours; When the failure probability of the photovoltaic power station equipment is less than 0.05, the operation and maintenance period of the photovoltaic power station equipment is set to 7 days.

8. An integrated photovoltaic power station multi-dimensional operation and maintenance management platform, characterized in that: The steps for implementing the multi-dimensional operation and maintenance management method of an integrated photovoltaic power station according to any one of claims 1 to 7, wherein the multi-dimensional operation and maintenance management platform of an integrated photovoltaic power station comprises: An operation mode acquisition module, the operation mode acquisition module is used to obtain the operation mode of the power station, wherein the operation mode of the power station has a starting photovoltaic power station equipment bit number list; A fault mining module, wherein the fault mining module is used to traverse the starting photovoltaic power station equipment bit number list to perform frequent pattern mining of fault samples based on the power station operation mode, and obtain a photovoltaic power station equipment fault probability list; A fault prediction module, the fault prediction module is used to traverse the starting photovoltaic power station equipment bit number list to perform fault prediction configuration and obtain a fault prediction model list; An operation and maintenance cycle determination module, the operation and maintenance cycle determination module is used to set a photovoltaic power station equipment operation and maintenance cycle list according to the photovoltaic power station equipment failure probability list; An operation and maintenance management module, wherein the operation and maintenance management module is used to perform operation and maintenance management of the photovoltaic power station according to the photovoltaic power station equipment operation and maintenance cycle list and the fault prediction model list.

Images