Photovoltaic power station multi-source fusion remote monitoring method and system based on cloud platform

Abstract

The invention discloses a photovoltaic power station multi-source fusion remote monitoring method and system based on a cloud platform, and relates to the technical field of photovoltaic power generation, and the method comprises the steps: deploying a plurality of data collection units at a target photovoltaic power station, collecting the multi-source operation data of the target photovoltaic power station in real time, uploading the multi-source operation data to a cloud data processing platform, and carrying out the fusion and correlation analysis, power generation performance deviation is identified, an abnormal cause is positioned, and a fault analysis result is generated; and generating a fault processing scheme according to a fault analysis result, and pushing the operation and maintenance work order to an operation and maintenance personnel terminal to execute field processing. The technical problems that the operation and maintenance efficiency is low and the power generation efficiency is limited due to the fact that operation and maintenance of an existing photovoltaic power station depend on manual inspection, the fault diagnosis period is long and the accuracy is poor are solved, multi-source fusion remote monitoring based on the cloud platform is achieved, real-time accurate diagnosis and quick response to faults of the photovoltaic power station are achieved, and the working efficiency of the photovoltaic power station is improved. And the operation and maintenance efficiency and the power generation benefit are improved.

Description

A Cloud-Based Method and System for Multi-Source Integration Remote Monitoring of Photovoltaic Power Plants Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, and more specifically to a method and system for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform. Background Technology

[0002] With the widespread application and continuous expansion of photovoltaic power plants, the importance of their operation and maintenance management is becoming increasingly prominent. Traditional photovoltaic power plant operation and maintenance mainly relies on manual inspections and experience-based judgment, which cannot obtain real-time equipment operating data. Furthermore, fault diagnosis depends on limited data and experience, which can easily lead to untimely fault detection and difficulty in accurately identifying the causes, increasing operation and maintenance time and costs. In addition, photovoltaic power plants are mostly located in remote areas, resulting in a large workload and low efficiency in on-site operation and maintenance, which affects the power plant's power generation efficiency and economic benefits. Summary of the Invention

[0003] This application provides a cloud-based method and system for multi-source fusion remote monitoring of photovoltaic power plants, which addresses the technical problems of existing photovoltaic power plant operation and maintenance relying on manual inspections, resulting in long fault diagnosis cycles and poor accuracy, leading to low operation and maintenance efficiency and limited power generation efficiency.

[0004] The first aspect of this application provides a cloud-based method for multi-source fusion remote monitoring of photovoltaic power plants. The method includes: deploying multiple data acquisition units at the target photovoltaic power plant to collect multi-source operational data in real time and uploading it to a cloud-based data processing platform; the cloud-based data processing platform performing fusion and correlation analysis on the received multi-source operational data to identify power generation performance deviations and locate the causes of anomalies; performing fault tracing analysis based on the causes of anomalies to generate fault analysis results; generating fault handling plans based on the fault analysis results and pushing maintenance work orders to the maintenance personnel's terminals; and the maintenance personnel performing on-site handling according to the maintenance work orders and feeding back the handling results to the cloud-based data processing platform, forming a closed-loop maintenance process.

[0005] The second aspect of this application provides a cloud-based photovoltaic power plant multi-source fusion remote monitoring system. The system includes: a multi-source operation data acquisition module, used to deploy multiple data acquisition units at the target photovoltaic power plant to collect multi-source operation data of the target photovoltaic power plant in real time and upload it to a cloud data processing platform; an anomaly identification and location module, used by the cloud data processing platform to perform fusion and correlation analysis on the received multi-source operation data, identify power generation performance deviations, and locate the causes of anomalies; a fault tracing and analysis module, used to perform fault tracing and analysis based on the causes of anomalies and generate fault analysis results; a maintenance work order push module, used to generate fault handling plans based on the fault analysis results and push maintenance work orders to the maintenance personnel's terminals; and a closed-loop maintenance management module, used by maintenance personnel to perform on-site handling according to the maintenance work orders and feed back the handling results to the cloud data processing platform, forming a closed-loop maintenance system.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application provides a cloud-based method and system for multi-source fusion remote monitoring of photovoltaic power plants, relating to the field of photovoltaic power generation technology. It achieves multi-source data fusion and remote monitoring of photovoltaic power plants through a cloud platform, combining intelligent fault diagnosis and closed-loop operation and maintenance management. This accurately identifies deviations in power generation performance and the causes of anomalies, quickly generates fault handling solutions, and pushes operation and maintenance work orders, enabling efficient execution and feedback of operation and maintenance work. It solves the technical problems of existing photovoltaic power plant operation and maintenance relying on manual inspections, resulting in long fault diagnosis cycles and poor accuracy, leading to low operation and maintenance efficiency and limited power generation efficiency. The system achieves multi-source fusion remote monitoring based on a cloud platform, enabling real-time accurate diagnosis and rapid response to photovoltaic power plant faults, thereby improving operation and maintenance efficiency and power generation benefits. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 is a schematic flowchart of a cloud-based photovoltaic power plant multi-source fusion remote monitoring method provided in an embodiment of this application;

[0010] Figure 2 is a schematic diagram of the structure of a cloud-based photovoltaic power plant multi-source fusion remote monitoring system provided in an embodiment of this application.

[0011] Figure labeling: 11 Multi-source operation data acquisition module, 12 Anomaly identification and location module, 13 Fault tracing and analysis module, 14 Operation and maintenance work order push module, 15 Closed-loop operation and maintenance management module. Detailed Implementation

[0012] This application provides a cloud-based method and system for multi-source fusion remote monitoring of photovoltaic power plants, which addresses the technical problems of existing photovoltaic power plant operation and maintenance relying on manual inspections, resulting in long fault diagnosis cycles and poor accuracy, leading to low operation and maintenance efficiency and limited power generation efficiency.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as shown in Figure 1, this application provides a multi-source fusion remote monitoring method for photovoltaic power plants based on a cloud platform, the method including:

[0016] P10: Deploy multiple data acquisition units at the target photovoltaic power station to collect multi-source operational data of the target photovoltaic power station in real time and upload it to the cloud data processing platform. The multi-source operational data includes at least the actual power generation parameters of the photovoltaic strings, inverter operating status data, meteorological environmental data, and component surface status data acquired by image acquisition equipment.

[0017] Specifically, to achieve comprehensive monitoring and analysis of the power plant's operational status, multiple data acquisition units need to be deployed at key locations within the power plant. These key locations include photovoltaic modules, combiner boxes, inverters, and weather stations, ensuring real-time acquisition of multi-source operational data from the power plant. Simultaneously, each data acquisition unit primarily consists of a series of sensors and image acquisition devices, capable of comprehensively collecting various types of operational data from the power plant. This data includes, but is not limited to, actual power generation parameters of the photovoltaic strings, inverter operating status data, meteorological environmental data, and module surface condition data.

[0018] Specifically, the actual power generation parameters of a photovoltaic (PV) string include key indicators such as current, voltage, and power, which reflect the real-time power generation performance of the PV modules. Inverter operating status data covers input and output current, voltage, power, temperature, and fault alarm information, used to assess the inverter's operational health. Meteorological environmental data includes irradiance, ambient temperature, wind speed, and humidity, which are crucial for analyzing the power generation efficiency and performance deviations of PV power plants. Furthermore, surface condition data of the modules acquired by image acquisition equipment can visually reflect whether there is dirt, shading, or physical damage on the PV module surface, providing important evidence for fault diagnosis.

[0019] To ensure data accuracy and reliability, the data acquisition unit transmits the collected data to the cloud data processing platform in real time using IoT technology. The application of IoT technology enables seamless communication between devices via wireless communication modules or wired network connections, thereby achieving efficient data transmission. To ensure the stability and security of data transmission, the data acquisition unit employs encryption protocols during transmission and uses data verification mechanisms to prevent data loss or tampering. Furthermore, the data acquisition unit must also have a certain data caching capacity to cope with network instability or interruptions, ensuring data integrity.

[0020] As the data hub of the entire system, the cloud-based data processing platform possesses powerful data processing and storage capabilities. Based on a cloud computing architecture, this platform can flexibly expand its resources according to the volume of data to meet the processing needs of large-scale data. After receiving multi-source operational data from various data acquisition units, the platform first performs data cleaning and formatting to remove invalid or abnormal data, ensuring data accuracy and consistency. Subsequently, the platform stores the processed data in a distributed database for subsequent data analysis and fault diagnosis.

[0021] In practical deployment, the selection and layout of data acquisition units are crucial. Data acquisition units for photovoltaic modules should possess high-precision current and voltage sensors to ensure the accuracy of measurement data; data acquisition units for combiner boxes must have multi-channel monitoring capabilities to simultaneously monitor the current and voltage of multiple branches; and data acquisition units for inverters must have high sampling rate and high-precision sensors to capture the inverter's operating status in real time. Data acquisition units for weather stations must be equipped with various environmental sensors to comprehensively monitor environmental parameters such as light intensity, temperature, wind speed, and humidity. By rationally arranging these data acquisition units, comprehensive coverage of all key areas of the photovoltaic power station can be ensured, thereby providing complete and accurate operational data to the cloud-based data processing platform.

[0022] P20: The cloud-based data processing platform integrates and correlates the received multi-source operating data to identify power generation performance deviations and pinpoint the causes of anomalies.

[0023] Furthermore, step P20 in this embodiment of the application also includes:

[0024] P21: Using meteorological environmental data from the multi-source operation data as input, the expected power generation of the target photovoltaic power station is generated through a photovoltaic power generation theoretical model; P22: The actual power generation parameters of the photovoltaic strings in the multi-source operation data are compared with the expected power generation to determine whether there is a power generation performance deviation; P23: When a power generation performance deviation is determined to exist, the component surface state image data and inverter operation state data of the period in which the deviation occurred and the corresponding photovoltaic array area are retrieved simultaneously for correlation analysis to locate the abnormal cause of the power generation performance deviation.

[0025] It should be understood that the cloud-based data processing platform is responsible for fusing and correlating the received multi-source operational data. These data sources are diverse, covering photovoltaic module output power, current, and voltage; combiner box current and voltage status; inverter operating parameters; and environmental data such as light intensity, ambient temperature, wind speed, and humidity collected by weather stations. The platform first standardizes this data to ensure consistency in format and units across different sources, thus providing a foundation for accurate analysis.

[0026] Subsequently, using meteorological and environmental data from multi-source operational data as input, the expected power generation of the target photovoltaic power station is generated through a photovoltaic power generation theoretical model. The photovoltaic power generation theoretical model is a deep learning or regression model based on physical principles, and its structure typically employs multilayer perceptrons (MLP) or convolutional neural networks (CNNs). The model's input layer receives various features from the meteorological and environmental data, the hidden layer performs feature extraction and nonlinear mapping through multiple neurons, and finally, the output layer generates the expected power generation of the target photovoltaic power station.

[0027] For example, the model receives input data including real-time collected meteorological environmental parameters such as light intensity, ambient temperature, wind speed, and humidity, as well as installation parameters of the photovoltaic power station, such as the tilt angle, azimuth angle, and geographical information of the installation location of the photovoltaic modules. Next, a feature extraction layer processes the input data to extract feature values ​​directly related to power generation. For example, it calculates the effective solar power of the photovoltaic modules based on light intensity and ambient temperature, and considers the impact of wind speed and humidity on module heat dissipation and power output. Then, in the power calculation layer, it simulates the output characteristics of the photovoltaic modules under current environmental conditions and calculates the expected power output of each photovoltaic module. Finally, in the system integration layer, the expected power output of individual photovoltaic modules is integrated into the system, considering the layout and connection methods of the entire photovoltaic power station, including the series and parallel connection methods of the photovoltaic modules, the conversion efficiency of the combiner box and inverter, etc., to derive the expected power generation of the entire photovoltaic power station.

[0028] After obtaining the expected power generation, the platform compares the actual power generation parameters of the photovoltaic strings in the multi-source operation data with the expected power generation. By calculating the difference between the actual and expected power generation, it determines whether there is a power generation performance deviation. If the actual power generation is lower than a certain threshold of the expected power generation, such as lower than 90% of the expected power, a power generation performance deviation is determined. This threshold can be adjusted based on the actual operation of the power plant and experience to ensure the accuracy of the deviation judgment.

[0029] When the platform determines that there is a deviation in power generation performance, it simultaneously retrieves image data of the photovoltaic module surface condition and the corresponding photovoltaic array area during the period of deviation, along with inverter operating status data, for correlation analysis to pinpoint the abnormal cause of the power generation performance deviation. For example, it acquires image data of the photovoltaic module surface condition within the corresponding time period from image acquisition equipment. This image data can intuitively reflect whether there are obstructions on the module surface, such as dust, bird droppings, or fallen leaves, and whether there is damage, such as cracks, hot spots, or other abnormalities. Through image recognition technology, the platform can automatically analyze the image data to identify potential abnormal factors. Simultaneously, the platform acquires inverter operating status data within the same time period, including input and output current, voltage, power, temperature, and fault alarm information. By analyzing the inverter's operating parameters, it determines whether there are equipment faults or operational abnormalities, such as inverter overheating, communication failures, or abnormal DC input.

[0030] Next, the surface condition image data of the modules is correlated with the inverter operating status data. By combining the differences between the actual power generation parameters and the expected power generation, the specific causes of the power generation performance deviation are comprehensively determined. For example, if the module surface image shows a large amount of dust obstruction, but the inverter is operating normally, it can be determined that the power generation performance deviation is mainly caused by dust on the module surface; if the inverter has a fault alarm, but the module surface condition is normal, it can be determined that the deviation is caused by an inverter fault.

[0031] Through this process, the cloud-based data processing platform can not only accurately identify deviations in power generation performance but also quickly pinpoint the abnormal causes of these deviations. This method effectively improves the fault detection efficiency of photovoltaic power plants, reduces the workload of operation and maintenance personnel, and ensures the stable and efficient operation of photovoltaic power plants under various environmental conditions.

[0032] Furthermore, in the association analysis, step P23 of this application embodiment also includes:

[0033] P23-1: Perform image recognition analysis on the surface condition data of the photovoltaic module to determine whether there is dirt, shading or physical damage on the surface of the photovoltaic module, and generate image recognition analysis results; P23-2: Analyze the temperature, input and output electrical parameters and alarm information in the inverter operating status data to perform inverter operating anomaly analysis and generate inverter operating status analysis results; P23-3: Based on the image recognition analysis results and the inverter operating status analysis results, locate the cause of the anomaly, the cause of the anomaly includes at least module surface anomaly, inverter equipment anomaly or a combination thereof.

[0034] Optionally, after the cloud data processing platform determines that there is a deviation in power generation performance, it can further perform detailed correlation analysis to locate the cause of the anomaly.

[0035] First, the platform performs image recognition analysis on the surface condition data of photovoltaic modules. This data comes from image acquisition equipment installed in photovoltaic power plants, which can acquire real-time images of the photovoltaic module surface condition. The image recognition analysis process uses computer vision technology, such as convolutional neural networks (CNNs), to automatically process this image data. First, it learns how to identify problems such as dirt, shading, or physical damage using training data, generating a recognition model. Then, it uses the real-time acquired photovoltaic module surface condition data as input to perform image recognition and generate image recognition analysis results. For example, if dust or foliage is detected on the photovoltaic module surface in the image, the model will mark these areas and generate image recognition analysis results for further analysis of their impact on power generation performance.

[0036] Secondly, the platform analyzes the inverter's operating status data, including temperature, input / output electrical parameters, and alarm information, to perform anomaly analysis. The inverter's operating status data includes real-time collected electrical parameters such as inverter operating temperature, input / output current, voltage, and power, as well as various alarm messages issued by the inverter. The platform analyzes these electrical parameters, combining them with the device's historical operating data, and uses status monitoring algorithms to assess whether the inverter's performance is abnormal. For example, if the inverter's operating temperature consistently exceeds the set normal range, or if abnormal fluctuations occur in the input / output electrical parameters, the platform will identify this as an inverter malfunction. Simultaneously, the analysis of alarm information directly reflects whether the inverter has faults or potential problems.

[0037] Finally, the platform comprehensively locates the cause of the anomaly based on the image recognition analysis results and the inverter operating status analysis results. The cause of the anomaly may include surface abnormalities of the photovoltaic modules (such as dirt, shading, or physical damage), inverter equipment abnormalities (such as excessively high temperature, abnormal electrical parameters, or fault alarms), or a combination of both. For example, if the image recognition results show dirt or shading on the surface of the photovoltaic modules, and the inverter is operating normally, the system can determine that the power generation performance problem originates from the photovoltaic modules themselves; if the inverter operating status is abnormal (such as excessively high temperature or unstable electrical parameters), the power generation performance problem is likely caused by an inverter fault; if both are present, the power generation performance deviation may be caused by a combination of surface abnormalities of the modules and inverter equipment abnormalities.

[0038] Through the above process, the cloud-based data processing platform can accurately locate the deviations in the power generation performance of photovoltaic power plants, providing detailed fault analysis for operation and maintenance personnel and ensuring that photovoltaic power plants operate more efficiently and stably.

[0039] P30: Based on the causes of the anomalies, conduct fault tracing analysis and generate fault analysis results.

[0040] Furthermore, step P30 in this embodiment of the application also includes:

[0041] P31: Using the aforementioned cause of the anomaly as input, query the preset fault diagnosis rule base, match and extract all associated potential fault points, and form a preliminary fault list; P32: Combining the historical change trend data of relevant electrical parameters in the multi-source operation data, traverse each potential fault point in the preliminary fault list, perform probability and urgency assessment, and generate fault assessment results; P33: Based on the fault assessment results, assign a fault level to each potential fault point, and integrate them to generate a comprehensive fault analysis result.

[0042] Specifically, after identifying the cause of the anomaly, the cloud-based data processing platform further performs fault tracing analysis to generate detailed fault analysis results, ensuring that maintenance personnel can take measures quickly and effectively.

[0043] First, the identified causes of anomalies are used as input to query a pre-defined fault diagnosis rule base. This rule base is a database containing various correlations between anomaly causes and potential fault points, covering various fault modes and their characteristic manifestations that may occur during the operation of photovoltaic power plants. Based on the anomaly cause, the platform queries and matches related potential fault points. Each fault point could be one of the causes of performance deviations, and the rule base provides the correlation and probability of occurrence for various faults based on past experience and data. In this way, the platform can extract fault points highly correlated with the anomaly causes from a large number of potential fault points, forming a preliminary fault list. The preliminary fault list lists all possible sources of faults, serving as the basis for further analysis and evaluation.

[0044] Subsequently, by combining historical trend data of relevant electrical parameters from multi-source operational data, an in-depth analysis is conducted on each potential fault point in the preliminary fault list. Specifically, the platform analyzes each potential fault point in the preliminary fault list one by one, examining their performance in historical operational data, such as the changing trends of electrical parameters like current, voltage, and power, to assess the probability and urgency of each potential fault point. For example, if a potential fault point involves inverter overheating, the platform analyzes the historical trend of inverter temperature changes and conducts further analysis in conjunction with other relevant electrical data. Based on historical trends, the probability and urgency of each fault point are assessed. The probability assessment mainly considers the frequency of the fault point occurring under similar conditions in the past or known fault modes, while the urgency assessment evaluates the immediate impact of the fault point on the power plant operation, such as whether it will lead to large-scale shutdowns or safety hazards. Based on these assessments, a fault assessment result is generated.

[0045] Finally, based on the above fault assessment results, a fault level is assigned to each potential fault point. Fault levels are typically categorized according to the severity of the fault, its probability of occurrence, and its impact on system operation, for example, into high, medium, and low levels. A high fault level indicates a high probability of fault occurrence and a significant impact on system operation, requiring priority handling; a low fault level indicates a low probability of fault occurrence or a minor impact on system operation, which can be temporarily observed or addressed later. Finally, by assigning a fault level to each potential fault point and combining it with the fault assessment results, a comprehensive fault analysis result is generated. This result not only includes specific information about the fault point but also provides priority suggestions for fault handling, helping operations and maintenance personnel accurately understand the severity and priority order of each potential fault point, thereby enabling more efficient fault diagnosis and repair.

[0046] Furthermore, by traversing each potential failure point in the preliminary failure list and assessing its probability and urgency, a failure assessment result is generated. Step P32 in this embodiment of the application further includes:

[0047] P32-1: For each potential fault point, retrieve historical trend data of its associated electrical parameters, calculate the deviation between the current parameter value and the historical normal benchmark value, assess the probability of the fault point occurring based on the deviation, and generate a fault probability assessment result; P32-2: Based on the fault type of each potential fault point and the contribution index of each potential fault point to the power plant's power generation, assess the maintenance urgency of the fault point and generate a fault urgency assessment result; P32-3: Weight and combine the fault probability assessment result and the fault urgency assessment result to generate the fault assessment result.

[0048] In one possible embodiment of this application, the fault assessment process is further refined by conducting a detailed probability and urgency assessment of each potential fault point to ensure that maintenance personnel are provided with comprehensive and accurate fault handling priorities.

[0049] First, for each potential fault point, historical trend data of its associated electrical parameters is retrieved. By analyzing this historical data, the platform can calculate the deviation between the current fault point's parameter value and historical normal baseline values. Baseline values ​​are typically based on verified stable or ideal state data from long-term power plant operation, representing the normal operating parameter range. For example, for inverter parameters such as voltage and temperature, historical baseline values ​​may be based on equipment manufacturer recommendations or long-term power plant operating data. The deviation can be calculated using statistical methods such as standard deviation and mean square error to assess the degree to which the current parameter deviates from the normal range. Based on this deviation, the system assesses the probability of the fault point occurring; the greater the difference between the current parameter value and the historical baseline value, the higher the probability of the fault point occurring. Finally, a fault probability assessment result is generated to quantify the probability of each potential fault point occurring.

[0050] Secondly, the platform assesses the maintenance urgency of each potential fault point based on its fault type and contribution index to the power plant's power generation. Different fault types mean different degrees of impact on power plant operation. For example, an inverter fault may have a direct impact on power generation, while minor contamination on the photovoltaic module surface has a relatively small impact. Simultaneously, the platform considers the contribution index of the fault point to the power plant's power generation. This contribution index can be determined based on the fault point's role in the overall power generation of the power plant; fault points with a higher contribution usually require priority repair. For fault points in critical equipment, even if their failure probability is not the highest, their maintenance urgency is still high due to their significant contribution to power generation. By combining fault type and contribution index, the platform can assess the urgency of the fault point—that is, its impact on the overall power plant's power generation efficiency, system safety, or equipment lifespan—generating a fault urgency assessment result and assigning a maintenance urgency level to each potential fault point.

[0051] Finally, the failure probability assessment results and failure urgency assessment results are weighted and synthesized. The weighting and synthesis process allocates weights based on the importance of each assessment result. Typically, the weights of the probability and urgency assessment results can be adjusted according to actual needs and the characteristics of the power plant's operation. For example, for power plants that prioritize power generation efficiency, the weight of maintenance urgency can be appropriately increased; while for power plants that aim to reduce the risk of sudden failures, the weight of failure probability can be increased. Through weighted synthesis, the platform generates a final failure assessment result for each potential failure point. This result not only reflects the probability of failure but also considers the urgency of maintenance, providing maintenance personnel with effective decision-making support, thereby ensuring efficient and timely failure handling and minimizing power plant downtime and losses.

[0052] P40: Based on the fault analysis results, generate a fault handling plan and push the maintenance work order to the maintenance personnel's terminal.

[0053] Furthermore, step P40 in this embodiment of the application also includes:

[0054] P41: Based on the potential fault points and fault levels in the fault analysis results, a fault handling knowledge base is matched to generate a fault handling plan; P42: The fault handling plan, the precise location information of the fault points in the fault analysis results, and the on-site data snapshot are automatically filled into the standard work order template to generate a standardized operation and maintenance work order; P43: Based on the area of ​​responsibility of the operation and maintenance personnel and the current work order load status, the standardized operation and maintenance work order is automatically allocated and pushed to the terminal of the most preferred operation and maintenance personnel, while triggering a processing time limit reminder.

[0055] Optionally, after completing the fault analysis and generating comprehensive fault analysis results, the cloud data processing platform enters the critical fault handling phase. Based on the fault analysis results, a targeted fault handling plan is generated, and relevant tasks are pushed to the maintenance personnel's terminals in the form of standardized maintenance work orders to ensure that the fault can be handled in a timely and effective manner.

[0056] First, the platform queries a pre-defined fault handling knowledge base based on the potential fault points and fault severity identified in the fault analysis results. This knowledge base is a database that centralizes various fault types and their corresponding handling methods, covering common causes of equipment failures, repair methods, component replacement suggestions, maintenance procedures, and other information. By analyzing the fault points and their severity (fault severity), the platform can accurately extract the most suitable handling solution from the knowledge base. For example, if the fault point is inverter overheating and the fault severity is high, the system might suggest checking the inverter's cooling system and performing necessary repairs or replacements. If the fault point involves dirt on the photovoltaic modules, the handling solution might be cleaning the modules or repairing related shading issues. Through this matching process, the platform can automatically generate personalized fault handling solutions based on different fault types.

[0057] Subsequently, the platform automatically populates the standard work order template with the generated fault handling plan, the precise fault location information from the fault analysis results, and the on-site data snapshot, generating a standardized maintenance work order. The precise fault location information includes the specific location of the fault, such as the photovoltaic module number and the inverter's installation location. The on-site data snapshot covers key operating parameters and environmental data at the time of the fault, such as the prevailing sunlight intensity, temperature, and module output power. This integration ensures that maintenance personnel have a comprehensive understanding of the fault situation before arriving on-site, thereby improving the efficiency and accuracy of fault handling. By combining this information with the fault handling plan, a standardized maintenance work order is generated, ensuring that the work order content is complete and structured. Maintenance personnel can directly execute operations based on the work order without needing to perform additional judgments or data searches.

[0058] Finally, the platform automatically allocates and pushes standardized maintenance work orders to the most suitable maintenance personnel's terminals based on the maintenance personnel's assigned area and current work order load status. The platform can intelligently select the most suitable maintenance personnel to handle the current work order by monitoring the maintenance personnel's work area and task load in real time. For example, if a maintenance personnel has fewer tasks and is close to the fault point, the work order will be prioritized for them. Simultaneously, a processing deadline reminder is triggered, notifying the maintenance personnel of the work order's urgency and suggested processing time, ensuring that maintenance personnel can respond to faults promptly and reduce the impact of faults on the operation of the photovoltaic power station. This automated work order allocation and deadline reminder function can greatly improve the efficiency of power station operation and maintenance, and can promptly handle various faults, ensuring the stable operation of the power station.

[0059] P50: The maintenance personnel perform on-site processing according to the maintenance work order and feed the processing results back to the cloud data processing platform, forming a closed loop of maintenance.

[0060] Specifically, after receiving a standardized maintenance work order, maintenance personnel will proceed to the fault site to perform the handling task based on the detailed information in the work order. The maintenance work order contains key information such as the fault handling plan, precise fault location information, and on-site data snapshots. This information provides clear guidance for maintenance personnel, ensuring that they can complete the fault handling work efficiently and accurately.

[0061] Upon arrival at the site, maintenance personnel first confirm the location of the fault and verify that the fault condition matches the work order description. Then, they execute the corresponding operations according to the handling plan provided in the work order. For example, if the fault is due to decreased power generation efficiency caused by dirt on the surface of the photovoltaic modules, the maintenance personnel will clean the module surface according to the cleaning plan; if the fault is due to inverter overheating, the maintenance personnel will inspect the inverter's cooling system, clean the heat dissipation channels, or replace damaged heat dissipation components.

[0062] After troubleshooting, maintenance personnel record the results in detail and feed them back to the cloud-based data processing platform. The feedback includes the specific measures taken to handle the fault, the processing time, changes in the on-site situation, and the post-troubleshooting equipment operating status. For example, maintenance personnel will record whether the surface of the photovoltaic modules is clean after cleaning and whether the inverter temperature has returned to normal. This real-time feedback ensures that the platform can continuously monitor the overall operating status of the power station and provide accurate data support for subsequent analysis and decision-making.

[0063] After receiving feedback from maintenance personnel, the cloud-based data processing platform analyzes and verifies the processing results, comparing equipment operating data before and after the process to confirm whether the fault has been completely resolved. If the fault handling is effective, the platform will close the work order and store the relevant data in the historical maintenance record for subsequent data analysis and maintenance optimization. If the fault is not completely resolved or new problems arise, the platform will regenerate the work order and continue to assign maintenance personnel to handle it until the problem is completely resolved.

[0064] Through on-site handling by maintenance personnel and feedback of the results, the cloud-based data processing platform achieves closed-loop management of maintenance work. This closed loop not only ensures that faults are resolved in a timely and effective manner, but also improves the overall efficiency and quality of photovoltaic power plant maintenance through continuous feedback and optimization mechanisms, providing a strong guarantee for the long-term stable operation of photovoltaic power plants.

[0065] In summary, the embodiments of this application have at least the following technical effects:

[0066] This application achieves real-time monitoring of photovoltaic power plants through multi-source data acquisition and intelligent analysis on a cloud platform, quickly identifying deviations in power generation performance and accurately locating the causes of faults, thus reducing troubleshooting time. Based on the fault analysis results, it provides intelligent fault tracing analysis and handling solutions to ensure accurate identification of fault causes and promotes efficient maintenance through standardized work orders. By automatically generating and allocating maintenance work orders, it enables reasonable scheduling of maintenance personnel and efficient task processing, shortening fault response time and improving the overall maintenance efficiency of the power plant. By feeding the processing results back to the cloud platform, it forms a closed-loop management system for maintenance, ensuring the traceability and effectiveness of each maintenance operation, and further optimizing the operation and maintenance process of the power plant.

[0067] It achieves the technical effect of multi-source fusion remote monitoring based on cloud platform, enabling real-time and accurate diagnosis and rapid response to photovoltaic power station faults, and improving operation and maintenance efficiency and power generation benefits.

[0068] Example 2: Based on the same inventive concept as the cloud-based photovoltaic power plant multi-source fusion remote monitoring method in the previous examples, as shown in Figure 2, this application provides a cloud-based photovoltaic power plant multi-source fusion remote monitoring system. The system and method examples in this application are based on the same inventive concept. The system includes:

[0069] The multi-source operation data acquisition module 11 is used to deploy multiple data acquisition units in the target photovoltaic power station to collect multi-source operation data of the target photovoltaic power station in real time and upload it to the cloud data processing platform.

[0070] The anomaly identification and location module 12 is used by the cloud data processing platform to fuse and correlate the received multi-source operating data, identify power generation performance deviations, and locate the causes of anomalies.

[0071] The fault tracing and analysis module 13 is used to perform fault tracing and analysis based on the cause of the anomaly and generate fault analysis results.

[0072] The maintenance work order push module 14 is used to generate a fault handling plan based on the fault analysis results and push the maintenance work order to the maintenance personnel's terminal.

[0073] The closed-loop operation and maintenance management module 15 is used by operation and maintenance personnel to perform on-site processing according to the operation and maintenance work order and to feed back the processing results to the cloud data processing platform, forming an operation and maintenance closed loop.

[0074] Furthermore, in the anomaly identification and location module 12:

[0075] The multi-source operating data includes at least the actual power generation parameters of the photovoltaic string, inverter operating status data, meteorological environment data, and component surface status data acquired by image acquisition equipment.

[0076] Furthermore, the anomaly identification and location module 12 is also used to perform the following steps:

[0077] Using meteorological environmental data from the multi-source operation data as input, the expected power generation of the target photovoltaic power station is generated through a photovoltaic power generation theoretical model. The actual power generation parameters of the photovoltaic strings in the multi-source operation data are compared with the expected power generation to determine whether there is a power generation performance deviation. When a power generation performance deviation is determined, the component surface state image data and inverter operation state data of the period in which the deviation occurred and the corresponding photovoltaic array area are retrieved simultaneously for correlation analysis to locate the abnormal cause of the power generation performance deviation.

[0078] Furthermore, the anomaly identification and location module 12 is also used to perform the following steps:

[0079] Image recognition analysis is performed on the surface condition data of the photovoltaic module to determine whether there is dirt, shading or physical damage on the surface of the photovoltaic module, and image recognition analysis results are generated; temperature, input and output electrical parameters and alarm information in the inverter operating status data are analyzed to perform inverter operating anomaly analysis and generate inverter operating status analysis results; based on the image recognition analysis results and the inverter operating status analysis results, the cause of the anomaly is located, and the cause of the anomaly includes at least module surface anomaly, inverter equipment anomaly or a combination thereof.

[0080] Furthermore, the fault tracing and analysis module 13 is also used to perform the following steps:

[0081] Using the aforementioned causes of the anomaly as input, a preset fault diagnosis rule base is queried to match and extract all associated potential fault points, forming a preliminary fault list. Combining the historical trend data of relevant electrical parameters in the multi-source operation data, each potential fault point in the preliminary fault list is traversed to assess its probability and urgency, generating a fault assessment result. Based on the fault assessment result, a fault level is assigned to each potential fault point, and the results are integrated to generate a comprehensive fault analysis result.

[0082] Furthermore, the fault tracing and analysis module 13 is also used to perform the following steps:

[0083] For each potential fault point, historical trend data of its associated electrical parameters are retrieved, the deviation between the current parameter value and the historical normal benchmark value is calculated, and the probability of the fault point occurring is assessed based on the deviation, generating a fault probability assessment result; based on the fault type of each potential fault point and the contribution index of each potential fault point to the power plant's power generation, the maintenance urgency of the fault point is assessed, generating a fault urgency assessment result; the fault probability assessment result and the fault urgency assessment result are weighted and combined to generate the fault assessment result.

[0084] Furthermore, the maintenance work order push module 14 is also used to perform the following steps:

[0085] Based on the potential fault points and fault levels in the fault analysis results, a fault handling plan is generated by matching a preset fault handling knowledge base. The fault handling plan, the precise location information of the fault points in the fault analysis results, and the on-site data snapshot are automatically filled into the standard work order template to generate a standardized operation and maintenance work order. Based on the area of ​​responsibility of the operation and maintenance personnel and the current work order load status, the standardized operation and maintenance work order is automatically allocated and pushed to the terminal of the most preferred operation and maintenance personnel, while triggering a processing time limit reminder.

[0086] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0087] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0088] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A cloud-based method for multi-source fusion remote monitoring of photovoltaic power plants, characterized in that, The method includes: deploying multiple data acquisition units at the target photovoltaic power station to collect multi-source operational data of the target photovoltaic power station in real time and uploading it to a cloud data processing platform; the cloud data processing platform performs fusion and correlation analysis on the received multi-source operational data to identify power generation performance deviations and locate the causes of anomalies; based on the causes of anomalies, it performs fault tracing analysis to generate fault analysis results; based on the fault analysis results, it generates a fault handling plan and pushes an operation and maintenance work order to the operation and maintenance personnel's terminal; the operation and maintenance personnel perform on-site handling according to the operation and maintenance work order and feed back the handling results to the cloud data processing platform, forming an operation and maintenance closed loop.

2. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 1, characterized in that, The multi-source operating data includes at least the actual power generation parameters of the photovoltaic string, inverter operating status data, meteorological environment data, and component surface status data acquired by image acquisition equipment.

3. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 1, characterized in that, The cloud-based data processing platform integrates and correlates the received multi-source operational data to identify power generation performance deviations and pinpoint the causes of anomalies. This includes: using meteorological environmental data from the multi-source operational data as input, generating the expected power generation of the target photovoltaic power station through a photovoltaic power generation theoretical model; comparing the actual power generation parameters of the photovoltaic strings in the multi-source operational data with the expected power generation to determine if there is a power generation performance deviation; and when a power generation performance deviation is determined to exist, simultaneously retrieving the component surface state image data and inverter operating state data for the period in which the deviation occurred and the corresponding photovoltaic array area, and performing correlation analysis to pinpoint the cause of the power generation performance deviation anomaly.

4. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 3, characterized in that, The correlation analysis includes: performing image recognition analysis on the surface condition data of the photovoltaic modules to determine whether there is dirt, shading, or physical damage on the surface of the photovoltaic modules, and generating image recognition analysis results; analyzing the temperature, input and output electrical parameters, and alarm information in the inverter operating status data to perform inverter operating anomaly analysis and generate inverter operating status analysis results; and locating the cause of the anomaly based on the image recognition analysis results and the inverter operating status analysis results, wherein the cause of the anomaly includes at least module surface anomalies, inverter equipment anomalies, or a combination thereof.

5. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 1, characterized in that, Based on the causes of the anomalies, a fault tracing analysis is performed to generate fault analysis results, including: taking the causes of the anomalies as input, querying a preset fault diagnosis rule base, matching and extracting all associated potential fault points to form a preliminary fault list; combining the historical change trend data of relevant electrical parameters in the multi-source operating data, traversing each potential fault point in the preliminary fault list, assessing its probability and urgency, and generating a fault assessment result; based on the fault assessment result, assigning a fault level to each potential fault point, and integrating them to generate a comprehensive fault analysis result.

6. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 5, characterized in that, The process involves iterating through each potential fault point in the preliminary fault list, assessing its probability and urgency, and generating a fault assessment result. This includes: for each potential fault point, retrieving historical trend data of its associated electrical parameters, calculating the deviation between the current parameter value and the historical normal baseline value, assessing the probability of the fault point occurring based on the deviation, and generating a fault probability assessment result; assessing the maintenance urgency of each potential fault point based on its fault type and its contribution index to the power plant's power generation, and generating a fault urgency assessment result; and weighting and combining the fault probability assessment result and the fault urgency assessment result to generate the final fault assessment result.

7. The method for multi-source fusion remote monitoring of photovoltaic power plants based on a cloud platform as described in claim 1, characterized in that, Based on the fault analysis results, a fault handling plan is generated, and an operation and maintenance work order is pushed to the operation and maintenance personnel's terminal. This includes: matching the potential fault points and fault levels in the fault analysis results with a preset fault handling knowledge base to generate a fault handling plan; automatically filling the fault handling plan, the precise fault point location information in the fault analysis results, and the on-site data snapshot into a standard work order template to generate a standardized operation and maintenance work order; and automatically allocating and pushing the standardized operation and maintenance work order to the most preferred operation and maintenance personnel's terminal based on the operation and maintenance personnel's responsible area and the current work order load status, while triggering a processing time limit reminder.

8. A cloud-based photovoltaic power plant multi-source fusion remote monitoring system, characterized in that, The system includes: a multi-source operation data acquisition module, used to deploy multiple data acquisition units at the target photovoltaic power station to collect multi-source operation data of the target photovoltaic power station in real time and upload it to the cloud data processing platform; an anomaly identification and location module, used by the cloud data processing platform to perform fusion and correlation analysis on the received multi-source operation data, identify power generation performance deviations and locate the causes of anomalies; a fault tracing and analysis module, used to perform fault tracing and analysis based on the causes of anomalies and generate fault analysis results; an operation and maintenance work order push module, used to generate fault handling solutions based on the fault analysis results and push operation and maintenance work orders to the operation and maintenance personnel's terminals; and a closed-loop operation and maintenance management module, used by operation and maintenance personnel to perform on-site handling according to the operation and maintenance work orders and feed back the handling results to the cloud data processing platform to form an operation and maintenance closed loop.

Images