Photovoltaic intelligent operation and maintenance monitoring big data management system
By utilizing the photovoltaic intelligent operation and maintenance monitoring big data management system, and employing digital twin and simulation analysis modules, the system solves the problems of manpower-intensive manual inspections and insufficient data analysis in traditional photovoltaic power plant operation and maintenance. It enables real-time monitoring and anomaly handling of photovoltaic power plant equipment status, thereby improving operation and maintenance efficiency and user satisfaction.
Patent Information
- Application Number
- CN202511826198.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-05
- Publication Date
- 2026-02-13
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional photovoltaic power plant operation and maintenance monitoring relies on manual inspections, which consumes a lot of manpower and resources, makes it difficult to grasp the equipment status in real time, and lacks in-depth mining and comprehensive analysis of massive amounts of data, resulting in insufficient prediction of equipment failures, increasing maintenance costs and power plant downtime.
The photovoltaic intelligent operation and maintenance monitoring big data management system is adopted. A digital twin model of the photovoltaic power station is established through the digital twin module. Combined with the simulation analysis module, real-time monitoring and simulation prediction are carried out to achieve accurate grasp of the status of power station equipment and abnormal handling.
It enables real-time and accurate monitoring of the operating status of photovoltaic power station equipment, improves operation and maintenance management efficiency, reduces economic losses caused by equipment failures, and enhances user experience.
Smart Images
Figure CN121529981A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of photovoltaic power plant operation and maintenance management technology, specifically a photovoltaic intelligent operation and maintenance monitoring big data management system. Background Technology
[0002] With the booming development of the photovoltaic power generation industry, photovoltaic power plants, as the core facilities for converting solar energy into electricity, are experiencing continuous expansion in installed capacity and increasingly widespread distribution, ranging from centralized large-scale ground-mounted power plants to decentralized industrial and commercial rooftop power plants and residential power plants. As the number and scale of photovoltaic power plants increase, the importance of operation and maintenance management becomes increasingly prominent. Efficient and accurate operation and maintenance monitoring has become key to ensuring stable operation of power plants, improving power generation efficiency, reducing operation and maintenance costs, and extending equipment lifespan.
[0003] Traditional photovoltaic power plant operation and maintenance monitoring methods mainly rely on regular manual inspections and simple data recording and analysis. Manual inspections not only require significant manpower and resources but also have long inspection cycles, making it difficult to monitor the real-time operating status of power plant equipment. Furthermore, manual inspections are susceptible to subjective influences, leading to errors in the results and failing to promptly identify potential equipment failures. Traditional methods often only record basic operating parameters, lacking in-depth data mining and comprehensive analysis of massive amounts of data, making it difficult to extract valuable information to guide operation and maintenance decisions. For example, the lack of systematic analysis of historical operating data and failure patterns makes it impossible to predict equipment failures in advance, resulting in reactive repairs after a failure occurs. This not only increases maintenance costs but also prolongs power outages, impacting power generation and economic benefits. Summary of the Invention
[0004] To address the problems existing in the above solutions, this invention provides a photovoltaic intelligent operation and maintenance monitoring big data management system.
[0005] The objective of this invention can be achieved through the following technical solutions: A photovoltaic intelligent operation and maintenance monitoring big data management system includes a platform terminal and a user terminal; The platform includes a digital twin module and a simulation analysis module; The digital twin module is used to create a digital twin model based on the user's photovoltaic power station.
[0006] Furthermore, the establishment of a digital twin model includes: Identify various potential photovoltaic power plants; classify these potential photovoltaic power plants to obtain their classifications. Select a potential photovoltaic power station from each power station category as a benchmark power station, create a digital twin based on the benchmark power station, and mark the digital twin as the benchmark twin; mark the benchmark twin with the corresponding power station category label; and store each digital twin. The system identifies the user's photovoltaic power station information, classifies the power station based on the information, matches the baseline twin of the power station classification, and adjusts the data based on the baseline twin to obtain a digital twin model of the user's photovoltaic power station.
[0007] Furthermore, various potential photovoltaic power plants are categorized, including: Step SA1: Calculate the market share of each potential photovoltaic power station, and sort the potential photovoltaic power stations in descending order of market share to obtain the first sequence; define the initial power station, which is the potential photovoltaic power station ranked first in the first sequence; The platform sets the classification criteria, which are such that the digital twin of the initial power station can be adjusted to be applicable to the corresponding potential photovoltaic power station, and the adjustment cost is no greater than the threshold X1. Step SA2: Identify the initial power station in real time according to the first sequence, determine whether the initial power station and each potential photovoltaic power station in the first sequence meet the classification criteria, and classify each potential photovoltaic power station and the initial power station that meet the classification criteria into one category and mark it as the power station classification; When there are no potential photovoltaic power plants that meet the classification criteria in the initial power plant, the initial power plant will be classified into one power plant category; Step SA3: Delete each potential photovoltaic power station corresponding to the power station category in the first sequence; Step SA4: Identify the number of potential photovoltaic power plants in the first sequence and mark them as the number of power plants; The analysis ends when the number of power stations equals 0. If the number of power stations is not equal to 0, return to step SA2.
[0008] Furthermore, a potential photovoltaic power station is selected as the benchmark power station from each power station category, including: Identify potential photovoltaic power plants corresponding to the initial power plant in the power plant classification, and mark the potential photovoltaic power plants as the benchmark power plants.
[0009] Furthermore, the various power station categories and benchmark power stations will be presented to users to guide them in building photovoltaic power stations according to the benchmark power stations.
[0010] Furthermore, before determining the benchmark power station, the various power stations are classified and screened, and the benchmark power station is determined based on the screened power station classification.
[0011] Furthermore, the various power plants are categorized and screened, including: Identify each potential photovoltaic power station corresponding to the power station category, and obtain the market share of the potential photovoltaic power station; sum the market shares of each potential photovoltaic power station to obtain the category share of the power station category; the platform presets the unit economic benefits; Identify a benchmark power station for power station classification and mark it as a reference power station; estimate the implementation cost of establishing a digital twin based on the reference power station; Substitute the implementation cost and category proportion into the preset screening formula to calculate the screening value for the power plant category. The screening formula is as follows: ; In the formula: TS is the screening value; CB is the implementation cost; β is the category proportion; DY is the unit economic benefit; Power plants with screening values below the threshold X2 are removed from the classification.
[0012] The simulation analysis module is used to simulate and analyze photovoltaic power plants, receive power plant monitoring data in real time, perform simulation prediction based on digital twin models and power plant monitoring data, and obtain operation simulation data of photovoltaic power plants; perform operation and maintenance analysis based on operation simulation data, obtain operation and maintenance management results, and send the operation and maintenance management results to the display module of the user terminal.
[0013] The user terminal includes a monitoring module, a power plant analysis module, and a display module; The monitoring module is used to monitor the photovoltaic power station in real time, obtain power station monitoring data, and send the power station monitoring data to the simulation analysis module on the platform. The display module is used to display the received data to the user.
[0014] Furthermore, the user terminal also includes a power plant analysis module; the simulation analysis module identifies whether there are operational anomalies based on the operation and maintenance management results; No action is taken when there are no operational or maintenance anomalies. When an operational anomaly occurs, the system identifies the anomaly based on the operational simulation data. Starting from the current time, it identifies the power plant monitoring data and the corresponding operational simulation data in real time based on the anomaly. The system then performs real-time calibration analysis on the operational simulation data based on the power plant monitoring data to obtain an initial calibration result, which includes calibration pass and calibration fail. Based on the initial calibration result, the system dynamically evaluates the anomaly trust value of the anomaly and sends the anomaly and the anomaly trust value to the power plant analysis module and display module on the user end. The power plant analysis module is used to analyze the operation of photovoltaic power plants and receive abnormal problems and abnormal trust values sent by the platform in real time. When the abnormal trust value is greater than the threshold X1, the abnormal problem is handled as an exception. When the abnormal trust value is not greater than the threshold X1, no abnormal problem is handled, and anomaly analysis is performed on the digital twin model.
[0015] Compared with the prior art, the beneficial effects of the present invention are: The solution proposed in this invention has significant beneficial effects in the field of photovoltaic power plant operation and maintenance monitoring. On the one hand, for the entire photovoltaic power generation industry, it breaks through the limitations of traditional operation and maintenance monitoring methods, no longer relying on a large investment of manpower and resources and manual periodic inspections that are greatly affected by subjective factors. It achieves real-time and accurate control of the operating status of photovoltaic power plant equipment, greatly improving the efficiency of operation and maintenance management. On the other hand, from the perspective of individuals and micro-users, this invention enables individuals and micro-users to manage their own photovoltaic power plants more easily and conveniently. Without the need for professional operation and maintenance knowledge and a large investment of time and energy, they can rely on this solution to achieve intelligent monitoring of the power plant, understand the operating status of the power plant in a timely manner, reduce economic losses caused by equipment failure, and improve the user experience and satisfaction of using photovoltaic power plants. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation
[0018] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0019] like Figure 1 As shown, a photovoltaic smart operation and maintenance monitoring big data management system includes a platform terminal and a user terminal; the platform terminal and the user terminal are connected by communication.
[0020] The platform includes a digital twin module and a simulation analysis module; The digital twin module is used to establish a corresponding digital twin model based on the user's photovoltaic power station.
[0021] In one embodiment, the digital twin model is built based on existing digital twin technology.
[0022] In one embodiment, for the numerous photovoltaic power plants currently used by individual users, the small scale and high cost of establishing digital twin models prevent many individual users from using them. Therefore, in this embodiment, the establishment of the digital twin model includes: Based on existing photovoltaic power station data from individuals or micro-enterprises, various types of photovoltaic power stations are identified. These stations are categorized based on factors such as installed capacity, component type, inverter type, and grid connection method, and marked as potential photovoltaic power stations. Each potential photovoltaic power station is then classified, with the classification criterion being that the digital twin model of a potential photovoltaic power station within the same category can be easily adjusted and applied to other potential photovoltaic power stations within the same category. The scope of this adjustment is set by the platform provider, and is evaluated based on factors such as adjustment cost and workload. This process yields the classification of each power station. Select a potential photovoltaic power station from each power station category as the benchmark power station, create a digital twin based on the benchmark power station, and label it as the benchmark twin; label the benchmark twin with the corresponding power station category tag; store each digital twin; Identifying a user's photovoltaic (PV) power plant information typically involves guiding the user to build a PV power plant according to various benchmark power plants during the initial setup process. This leads to higher accuracy and lower costs in subsequent analysis. Based on the PV power plant information, the corresponding power plant category is identified, and a benchmark twin for that category is determined. Adjustments are then made based on the benchmark twin to obtain a digital twin model of the user's PV power plant. In other words, the benchmark twin is adjusted according to the user's actual PV power plant, and further adjustments are made based on environmental and other differences to obtain a digital twin model of the user that meets the required accuracy.
[0023] In one embodiment, the classification of various potential photovoltaic power plants can be based on existing clustering algorithms or other classification methods.
[0024] In one embodiment, various potential photovoltaic power plants are categorized, including: Step SA1: Calculate the market share of each potential photovoltaic power station, and sort the potential photovoltaic power stations in descending order of market share to obtain the first sequence; define the initial power station, which is the potential photovoltaic power station ranked first in the first sequence; the platform sets the classification criteria, which are that the digital twin of the initial power station can be adjusted to be applicable to the corresponding potential photovoltaic power station, and the adjustment cost is not greater than the threshold X1. The initial power station and the potential photovoltaic power station are considered to meet the classification criteria; however, this does not mean that the reverse is necessarily true.
[0025] Step SA2: Identify the initial power station in real time according to the first sequence, determine whether the initial power station and each potential photovoltaic power station in the first sequence meet the classification criteria, and classify each potential photovoltaic power station and the initial power station that meet the classification criteria into one category and mark it as the power station classification; When the initial power station does not have any potential photovoltaic power stations that meet the classification criteria, the initial power station is classified independently. Step SA3: Delete each potential photovoltaic power station corresponding to the power station category in the first sequence, including the initial power station corresponding to that power station category; Step SA4: Identify the number of potential photovoltaic power plants in the first sequence and mark them as the number of power plants; The analysis ends when the number of power stations equals 0. If the number of power stations is not equal to 0, return to step SA2.
[0026] In one embodiment, determining whether the initial power station meets the classification criteria with each potential photovoltaic power station in the first sequence can be estimated based on existing methods, such as manual judgment, or it can be based on intelligent models built using machine learning, deep learning algorithms, etc., to determine the corresponding adjustment costs, and then compare and judge according to the threshold X1.
[0027] In one embodiment, a potential photovoltaic power station is selected as the benchmark power station from each power station category. The selection is based on the specific classification method. For example, if the classification is carried out according to the above embodiment, the power station with the highest market share or the original initial power station is selected. If the classification is based on other classification methods, the selection can also be based on factors such as cost.
[0028] In one embodiment, to protect the platform's revenue, when various power station categories are obtained, each power station category is filtered, and a benchmark power station is selected from the filtered power station categories.
[0029] In one embodiment, the various power station categories are filtered, which can be based on existing methods to eliminate power station categories that do not meet the platform's profit targets.
[0030] In one embodiment, filtering is performed on the various power plant categories, including: Identify each potential photovoltaic power station corresponding to a power station category and obtain the market share of the potential photovoltaic power stations; sum the market shares of each potential photovoltaic power station to obtain the category share of the power station category; the platform presets the economic benefits corresponding to different unit market shares and marks them as unit economic benefits; The benchmark power station, which is classified according to the method for determining the benchmark power station, is marked as the reference power station; the estimated cost of establishing a digital twin based on the reference power station is marked as the implementation cost. Substitute the implementation cost and category proportion into the preset screening formula to calculate the screening value for the power plant category. The screening formula is as follows: ; In the formula: TS is the screening value; CB is the implementation cost; β is the category proportion; DY is the unit economic benefit; Power plants with screening values below the threshold X2 are removed from the classification.
[0031] In one embodiment, the screening formula can also take other forms, such as considering different proportional coefficients, revenue corrections, etc., because the unit economic returns do not take into account the differences between different benchmark power plants. Therefore, corresponding correction coefficients can be set to correct the unit economic returns and improve the calculation accuracy.
[0032] In one embodiment, filtering can also be performed directly based on the category percentage, that is, power plant categories with a category percentage lower than a preset value are eliminated.
[0033] The simulation analysis module is used to simulate and analyze photovoltaic power plants, receive power plant monitoring data in real time, perform simulation predictions based on digital twin models and power plant monitoring data, and obtain simulated operation data of photovoltaic power plants for a period of time in the future; perform operation and maintenance analysis based on the operation simulation data, obtain operation and maintenance management results, send the operation and maintenance management results to the display module of the user terminal; and identify whether there are operation and maintenance anomalies based on the operation and maintenance management results. Operation and maintenance anomalies refer to abnormal conditions of photovoltaic power plants, such as various faults. No action is taken when there are no operational or maintenance anomalies. When an operational anomaly occurs, the system identifies the anomaly based on operational simulation data. Starting from the current time, it identifies the corresponding power plant monitoring data and operational simulation data for the same time period. This includes the power plant monitoring data and operational simulation data related to the anomaly. The system then performs real-time calibration analysis on the operational simulation data based on the power plant monitoring data to obtain initial calibration results, which include calibration pass and calibration fail. Based on the initial calibration results, the system dynamically assesses the anomaly trust value and sends the anomaly and its trust value to the power plant analysis and display modules on the user's end.
[0034] In one embodiment, the operation simulation data is calibrated and analyzed in real time based on the power plant monitoring data. That is, the deviation of the operation simulation data is evaluated based on the power plant monitoring data to see if it exceeds the preset standard. If it does not exceed the standard, it is considered to be calibrated as qualified; otherwise, it is considered to be calibrated as unqualified.
[0035] In one embodiment, the anomaly trust value of the abnormal problem is dynamically evaluated based on the initial calibration results. The credibility value of the operational simulation data for the abnormal problem is determined based on the deviation between the power plant monitoring data and the operational simulation data, and marked as the anomaly trust value. Existing credibility value evaluation methods are used for this assessment. For example, the anomaly trust value is calculated based on the proportion of subsequent calibrations that pass and fail. Adjustments can also be made based on the degree to which the calibration fails and exceeds the standard. There are various ways to calculate the anomaly trust value; intelligent evaluation can also be performed using intelligent models built based on machine learning, deep learning algorithms, etc.
[0036] In one embodiment, relevant correlation verification data can also be collected from the photovoltaic power station based on the abnormal problem, that is, which photovoltaic power station data is used to verify the cause of the abnormality; the cause of the abnormality is verified based on the correlation verification data, such as shading problem or equipment aging problem; Example: Component hot spot effect verification: Model prediction: The digital twin model simulation found that the output power of a certain component string was lower than expected and the temperature distribution was uneven, predicting the presence of a hot spot effect.
[0037] Related verification data: Infrared thermal imaging: Actual detection showed that the surface temperature of a certain component in the component string was 20°C higher than the surrounding area, confirming a hot spot.
[0038] EL testing revealed a microcrack in the component, causing localized current blockage.
[0039] Conclusion: The model prediction is accurate; the faulty component needs to be replaced.
[0040] Verification of inverter efficiency degradation: Model prediction: The model shows that the conversion efficiency of a certain inverter drops from 98% to 95%, which is below the threshold.
[0041] Related verification data: Actual efficiency calculation: The actual efficiency was calculated using the input / output power ratio, and it was confirmed to have dropped to 94.8%.
[0042] Cooling fan speed: An abnormally low fan speed was detected, causing the internal temperature of the inverter to rise.
[0043] Conclusion: The inverter's cooling system is faulty; the fan needs to be cleaned or the bearings replaced.
[0044] Verification of the effect of shadow occlusion: Model prediction: The model simulation shows that the output power of the components in a certain area decreased by 15% in the afternoon, which is predicted to be due to the shadow of nearby buildings.
[0045] Validation data: Light sensor: The actual light intensity drops to 80% of the model input value during the prediction period.
[0046] Aerial drone footage: Images show building projections covering components in the area.
[0047] Conclusion: Shadow occlusion is confirmed; component layout needs to be adjusted or tracking brackets need to be installed.
[0048] The user terminal includes a monitoring module, a power plant analysis module, and a display module; The monitoring module is used to monitor the photovoltaic power station in real time based on the data required for simulation by the digital twin model and subsequent verification data, obtain the corresponding power station monitoring data, and send the power station monitoring data to the simulation analysis module on the platform.
[0049] The power plant analysis module is used to analyze the operation of photovoltaic power plants and receive abnormal problems and abnormal trust values sent by the platform in real time. When the abnormal trust value is greater than the threshold X1, the abnormal problem is handled as an exception. When the abnormal trust value is not greater than the threshold X1, no abnormal problem is handled, and anomaly analysis is performed on the digital twin model; that is, when an analysis error is determined, the cause of the error is determined, and then the digital twin model is optimized and adjusted according to the cause of the error.
[0050] In one embodiment, abnormal problems are handled in accordance with existing abnormal problem handling methods, such as displaying warnings to users through a display module and isolating abnormal devices.
[0051] In one embodiment, abnormal problems can be handled according to their severity and timeliness. If the problem is not urgent, on-site verification can be conducted. If time is of the essence, emergency measures can be taken immediately.
[0052] The display module is used to display the received data to the user.
[0053] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.
[0054] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.
Claims
1. A photovoltaic intelligent operation and maintenance monitoring big data management system, characterized in that, Including both the platform side and the user side; The platform includes a digital twin module and a simulation analysis module; The digital twin module is used to build a digital twin model based on the user's photovoltaic power station; The simulation analysis module is used to simulate and analyze photovoltaic power plants, receive power plant monitoring data in real time, perform simulation and prediction based on digital twin models and power plant monitoring data, and obtain operation simulation data of photovoltaic power plants; perform operation and maintenance analysis based on operation simulation data, obtain operation and maintenance management results, and send the operation and maintenance management results to the display module of the user terminal; The user terminal includes a monitoring module and a display module; The monitoring module is used to monitor the photovoltaic power station in real time, obtain power station monitoring data, and send the power station monitoring data to the simulation analysis module on the platform. The display module is used to display the received data to the user.
2. The photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 1, characterized in that, The establishment of a digital twin model includes: Identify various potential photovoltaic power plants; classify these potential photovoltaic power plants to obtain their classifications. Select a potential photovoltaic power station from each power station category as a benchmark power station, create a digital twin based on the benchmark power station, and mark the digital twin as the benchmark twin; label the benchmark twin with the corresponding power station category tag; and store each digital twin. The system identifies the user's photovoltaic power station information, classifies the power station based on the information, matches a baseline twin of the power station classification, and adjusts the data based on the baseline twin to obtain a digital twin model of the user's photovoltaic power station.
3. The photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 2, characterized in that, Various potential photovoltaic power plants are categorized, including: Step SA1: Calculate the market share of each potential photovoltaic power station, sort the potential photovoltaic power stations in descending order of market share to obtain the first sequence; define the initial power station, which is the potential photovoltaic power station ranked first in the first sequence; The platform sets the classification criteria, which are such that the digital twin of the initial power station can be adjusted to be applicable to the corresponding potential photovoltaic power station, and the adjustment cost is no greater than the threshold X1. Step SA2: Identify the initial power station in real time according to the first sequence, determine whether the initial power station and each potential photovoltaic power station in the first sequence meet the classification criteria, and classify each potential photovoltaic power station and the initial power station that meet the classification criteria into one category and mark it as the power station classification; When there are no potential photovoltaic power plants that meet the classification criteria in the initial power plant, the initial power plant will be classified into one power plant category; Step SA3: Delete each potential photovoltaic power station corresponding to the power station category in the first sequence; Step SA4: Identify the number of potential photovoltaic power plants in the first sequence and mark them as the number of power plants; The analysis ends when the number of power stations equals 0. If the number of power stations is not equal to 0, return to step SA2.
4. The photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 3, characterized in that, Select one potential photovoltaic power plant from each power plant category as the benchmark power plant, including: Identify potential photovoltaic power plants corresponding to the initial power plant in the power plant classification, and mark the potential photovoltaic power plants as the benchmark power plants.
5. A photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 2, characterized in that, The system categorizes and displays benchmark power stations to users, guiding them to build photovoltaic power stations according to these benchmarks.
6. A photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 2, characterized in that, Before determining the benchmark power station, the various power stations are classified and screened, and the benchmark power station is determined based on the screened power station classification.
7. A photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 6, characterized in that, The power plants were categorized and screened, including: Identify each potential photovoltaic power station corresponding to the power station category, and obtain the market share of the potential photovoltaic power station; sum the market shares of each potential photovoltaic power station to obtain the category share of the power station category; the platform presets the unit economic benefits; Identify a benchmark power station for power station classification and mark it as a reference power station; estimate the implementation cost of establishing a digital twin based on the reference power station; Substitute the implementation cost and category proportion into the preset screening formula to calculate the screening value for the power plant category. The screening formula is as follows: ； In the formula: TS is the screening value; CB is the implementation cost; β is the category proportion; DY is the unit economic benefit; Power plants with screening values below the threshold X2 are removed from the classification.
8. A photovoltaic intelligent operation and maintenance monitoring big data management system according to claim 1, characterized in that, The user terminal also includes a power plant analysis module; the simulation analysis module identifies whether there are operation and maintenance anomalies based on the operation and maintenance management results. No action is taken when there are no operational or maintenance anomalies. When an operational anomaly occurs, the system identifies the anomaly based on the operational simulation data. Starting from the current time, it identifies the power plant monitoring data and the corresponding operational simulation data in real time based on the anomaly. The system then performs real-time calibration analysis on the operational simulation data based on the power plant monitoring data to obtain an initial calibration result, which includes calibration pass and calibration fail. Based on the initial calibration result, the system dynamically evaluates the anomaly trust value of the anomaly and sends the anomaly and the anomaly trust value to the power plant analysis module and display module on the user end. The power plant analysis module is used to analyze the operation of photovoltaic power plants and receive abnormal problems and abnormal trust values sent by the platform in real time. When the abnormal trust value is greater than the threshold X1, the abnormal problem is handled as an exception. When the abnormal trust value is not greater than the threshold X1, no abnormal problem is handled, and anomaly analysis is performed on the digital twin model.