Photovoltaic power station operation fault detection system based on data acquisition and analysis

Abstract

The invention belongs to the field of photovoltaic fault detection, relates to a data analysis technology, is used for solving the problem that a targeted operation and maintenance decision cannot be generated by combining an overall operation state evaluation result in the prior art, and particularly relates to a photovoltaic power station operation fault detection system based on data acquisition and analysis. Comprising a region detection module, an operation and maintenance evaluation module and an operation and maintenance analysis module which are connected in sequence, and the region detection module, the operation and maintenance evaluation module and the operation and maintenance analysis module are in communication connection with a database; the region detection module is used for carrying out regional detection analysis on the photovoltaic power station; comprehensive detection and intelligent operation and maintenance of the operation state of the photovoltaic power station are realized, the system can accurately position the fault position and type through region division and multi-parameter acquisition, the system can capture a cumulative abnormal trend through a periodic evaluation mechanism, and on the whole, the system remarkably improves the operation reliability of the photovoltaic power station, and the system is suitable for popularization and application. And powerful support is provided for intelligent development of the photovoltaic industry.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic fault detection and involves data analysis technology. Specifically, it is a photovoltaic power plant operation fault detection system based on data acquisition and analysis. Background Technology

[0002] Traditional photovoltaic power plant operation and maintenance mainly rely on manual inspections and regular maintenance, which are inefficient, costly, and difficult to detect hidden faults. The photovoltaic power plant operation fault detection system collects photovoltaic power plant operation data in a comprehensive and real-time manner, and uses big data, Internet of Things and artificial intelligence technologies to build an intelligent and automated fault detection and diagnosis platform.

[0003] The invention patent with publication number CN118944593B discloses a photovoltaic power station fault detection method based on knowledge graphs. This method transforms heterogeneous time-series data into structured graph data by constructing a knowledge graph. Based on the knowledge graph, a graph neural network model is built and node features are learned to generate embedded representations of nodes. A fault detection model is established and fault identification is achieved through link prediction technology, enabling the system to identify potential faults in the early stages, reduce downtime of photovoltaic power stations, and improve operating efficiency. However, this method cannot perform overall operating status assessment and analysis based on periodic detection results, nor can it generate targeted operation and maintenance decisions based on the overall operating status assessment results. This results in low operation and maintenance efficiency of photovoltaic power stations, and the implicit risks accumulated by various detection parameters over a period of time cannot be fed back.

[0004] To address the aforementioned technical problems, this application proposes a solution. Summary of the Invention

[0005] The purpose of this invention is to provide a photovoltaic power plant operation fault detection system based on data acquisition and analysis, which solves the problem that existing technologies cannot generate targeted operation and maintenance decisions by combining the overall operation status assessment results; The technical problem to be solved by this invention is: how to provide a photovoltaic power plant operation fault detection system based on data acquisition and analysis that can generate targeted operation and maintenance decisions by combining the overall operation status assessment results.

[0006] The objective of this invention can be achieved through the following technical solutions: A photovoltaic power plant operation fault detection system based on data acquisition and analysis includes a regional detection module, an operation and maintenance assessment module, and an operation and maintenance analysis module connected in sequence. The regional detection module, the operation and maintenance assessment module, and the operation and maintenance analysis module are all communicatively connected to a database. The regional detection module is used to perform regional detection and analysis on the photovoltaic power station: the photovoltaic power station is divided into detection areas i, i=1, 2, ..., n, where n is a positive integer; sensors are deployed according to the equipment in the detection area and parameters e, e=1, 2, ..., m, where m is a positive integer, a detection cycle is generated, and several detection time points with equal time intervals are set within the detection cycle. Fault detection is performed at the detection time points and the detection time points are marked as normal time points or abnormal time points. The operation and maintenance assessment module is used to perform overall operation and maintenance assessment and analysis of the photovoltaic power station: obtain the abnormal performance value and fault performance value of the detection period, mark the sum of the abnormal performance value and fault performance value as the operation and maintenance requirement value, and determine whether the overall operating status of the photovoltaic power station meets the requirements within the detection period through the operation and maintenance requirement value. The operation and maintenance analysis module is used to perform operation and maintenance decision analysis for photovoltaic power plants.

[0007] Furthermore, the specific process of fault detection at the detection time point includes: collecting the value of parameter e in the detection area i at the detection time point and marking it as the collected value CJie; retrieving the standard value BZie corresponding to parameter e; marking the absolute value of the difference between the collected value CJie and the standard value BZie as the detection value JCie of parameter e; and marking the detection area as a normal area or an abnormal area through the detection value JCie.

[0008] Furthermore, the specific process of marking the detection area as a normal or abnormal area includes: retrieving the detection threshold JCed of parameter e from the database, comparing the detection values ​​JCie of all parameters e within detection area i with the corresponding detection threshold JCed; if all detection values ​​JCie are less than the corresponding detection threshold JCed, the fault detection result of the detection area is determined to meet the requirements, and the corresponding detection area is marked as a normal area; otherwise, the fault detection result of the detection area is determined to not meet the requirements, the corresponding detection area is marked as an abnormal area, a fault processing signal is generated, and the fault processing signal is sent to the mobile terminal of the management personnel.

[0009] Furthermore, the specific process of marking the detection time point as a normal time point or an abnormal time point includes: if all detection areas are marked as normal areas, then the corresponding detection time point is marked as a normal time point; otherwise, the corresponding detection time point is marked as an abnormal time point.

[0010] Furthermore, the process of obtaining abnormal performance values ​​and fault performance values ​​includes: marking the ratio of the number of abnormal areas to the number of normal areas corresponding to the detection time point as the abnormal coefficient; marking the maximum value of the abnormal coefficient of all detection time points within the detection period as the abnormal performance value; and marking the ratio of the number of times abnormal time points are marked to the number of times normal time points are marked within the detection period as the fault performance value.

[0011] Furthermore, the specific process for determining whether the overall operating status of a photovoltaic power station meets the requirements during the testing period includes: retrieving the operation and maintenance requirement threshold from the database and comparing the operation and maintenance requirement value with the operation and maintenance requirement threshold; if the operation and maintenance requirement value is less than the operation and maintenance requirement threshold, it is determined that the overall operating status of the photovoltaic power station meets the requirements during the testing period and does not have operation and maintenance characteristics; if the operation and maintenance requirement value is greater than or equal to the operation and maintenance requirement threshold, it is determined that the overall operating status of the photovoltaic power station does not meet the requirements during the testing period and has operation and maintenance characteristics, generating an operation and maintenance analysis signal and sending the operation and maintenance analysis signal to the operation and maintenance analysis module.

[0012] Furthermore, the specific process of the operation and maintenance analysis module for performing operation and maintenance decision analysis on photovoltaic power plants includes: obtaining the decision coefficient for the detection cycle, obtaining the decision threshold through the database, and comparing the decision coefficient with the decision threshold; if the decision coefficient is less than the decision threshold, a comprehensive maintenance signal is generated and sent to the mobile terminal of the management personnel; if the decision coefficient is greater than or equal to the decision threshold, the L1 parameters e with the largest out-of-standard performance values ​​are marked as optimization objects, a key optimization signal is generated, and the key optimization signal and the optimization objects are sent to the mobile terminal of the management personnel.

[0013] Furthermore, the process of obtaining the decision coefficients for the detection cycle includes: marking the parameter e where the detection value JCie at the abnormal time point is not less than the corresponding detection threshold JCed as the analysis object; marking the ratio of the detection value JCie of the analysis object at the abnormal time point to the detection threshold JCed as the exceeding value of the analysis object; marking the maximum value of all parameters e as the exceeding performance value of the analysis object; and calculating the variance of all exceeding performance values ​​of parameters e to obtain the decision coefficients.

[0014] The present invention has the following beneficial effects: This application enables comprehensive monitoring and intelligent operation and maintenance of photovoltaic power plants. Through regional division and multi-parameter acquisition, the system can accurately locate the fault location and type. The periodic evaluation mechanism allows the system to capture cumulative abnormal trends, rather than relying solely on data from a single point in time. Quantified operation and maintenance requirements provide an objective basis for operation and maintenance decisions, avoiding the subjectivity and lag of human judgment. Targeted operation and maintenance suggestions help improve maintenance efficiency and reduce operation and maintenance costs. Overall, the system significantly improves the operational reliability and economic benefits of photovoltaic power plants, providing strong support for the intelligent development of the photovoltaic industry. This application enables precise quantitative detection of parameters in each detection area of ​​a photovoltaic power station. By comparing the collected values ​​with standard values, the detected values ​​are calculated, providing an objective basis for subsequent judgment of the area status. This method avoids errors that may be caused by subjective human judgment, improves the accuracy and reliability of fault detection, and, due to the adoption of a numerical detection method, makes the detection results of different parameters and different areas comparable, which facilitates unified evaluation and analysis. This application enables a quantitative assessment of the operating status of photovoltaic power plants. As a result, the system can accurately reflect the abnormal and fault conditions of photovoltaic power plants during the detection period, providing reliable data support for subsequent operation and maintenance decisions. This quantitative assessment method avoids the bias of subjective judgment and improves the accuracy and reliability of fault detection. At the same time, by calculating abnormal and fault conditions separately, the system can more comprehensively reflect the operating status of photovoltaic power plants, helping operation and maintenance personnel to more accurately identify and handle potential problems. 4. This application enables accurate assessment of the operating status of photovoltaic power plants and targeted operation and maintenance decisions. By introducing a comparison mechanism of decision coefficients and decision thresholds, the system can automatically determine whether comprehensive maintenance or key optimization is needed based on the actual operating conditions. When key optimization is needed, the system identifies the parameter with the highest out-of-standard performance value as the optimization target, allowing operation and maintenance work to focus on the aspects that need the most improvement. This method not only improves the accuracy and efficiency of operation and maintenance but also effectively reduces operation and maintenance costs and extends the service life of photovoltaic power plants. At the same time, by sending operation and maintenance information directly to the mobile terminals of management personnel, rapid response and timely processing are achieved, minimizing the downtime and power generation loss of photovoltaic power plants. Attached Figure Description

[0015] 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.

[0016] Figure 1 This is a system block diagram of Embodiment 1 of the present invention; Figure 2 This is a flowchart of the method in Embodiment 2 of the present invention. Detailed Implementation

[0017] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] In the traditional operation and maintenance (O&M) of photovoltaic power plants, the reliance on manual inspections and periodic maintenance is inefficient and costly, and particularly difficult to identify hidden faults. Although knowledge graph-based fault detection methods can achieve early fault identification through graph neural networks, they lack the ability to comprehensively evaluate periodic detection data. They cannot combine dynamic indicators such as the distribution of abnormal time points, the proportion of regional anomalies, and the degree of parameter exceedance to generate differentiated O&M strategies. This results in the system being unable to effectively balance the allocation of maintenance resources and the control of hidden risks, leading to delayed O&M response and the accumulation of parameter anomalies.

[0019] For example, in photovoltaic power plants using traditional detection methods, the system can only judge the equipment status based on isolated data from a single point in time. It cannot establish a dynamic evaluation model based on indicators such as changes in the number of abnormal areas at multiple time points within the detection period, or the degree of continuous deviation between parameter detection values ​​and thresholds. When some parameters in a certain detection area slightly exceed the standard at multiple consecutive time points, the system cannot identify the growth trend of the abnormal coefficient in that area, nor can it correlate the differences in abnormal coefficients between different areas with maintenance priorities. Furthermore, because there is no correlation analysis mechanism between the density of abnormal time points and the degree of parameter exceeding the standard within the detection period, maintenance personnel find it difficult to determine whether to implement comprehensive maintenance or optimize specific parameters, leading to wasted maintenance resources or failure to promptly eliminate risks associated with critical parameters.

[0020] If the above problems are not addressed, abnormal patterns and parameter correlation features in periodic monitoring data cannot be effectively extracted. System operation and maintenance decisions will rely on empirical judgment for a long time, making it difficult to avoid accelerated component performance degradation caused by persistent anomalies in local parameters. The accumulation of latent faults may trigger a chain reaction in multiple regions, increasing the risk of overall power plant shutdown. At the same time, repetitive maintenance operations will significantly increase labor and material costs, ultimately affecting the stability and economic benefits of power plant operation.

[0021] Example 1: As Figure 1 As shown, a photovoltaic power plant operation fault detection system based on data acquisition and analysis includes a regional detection module, an operation and maintenance assessment module, and an operation and maintenance analysis module connected in sequence. The regional detection module, the operation and maintenance assessment module, and the operation and maintenance analysis module are all connected to a database.

[0022] The regional detection module is used for regional detection and analysis of photovoltaic power plants. It divides the photovoltaic power plant into detection regions i, i = 1, 2, ..., n, where n is a positive integer. Sensors are deployed according to the equipment within each detection region, and parameters e, e = 1, 2, ..., m, where m is a positive integer, are generated. A detection cycle is generated, and several equally timed detection points are set within the detection cycle. At each detection point, the value of parameter e within detection region i is collected and marked as the collected value CJie. The corresponding standard value BZie for parameter e is retrieved. The absolute value of the difference between the collected value CJie and the standard value BZie is marked as the detection value JCie for parameter e. The detection threshold JC for parameter e is retrieved from the database. The system compares the detection values ​​JCie of all parameters e within detection area i with the corresponding detection threshold JCed. If all detection values ​​JCie are less than the corresponding detection threshold JCed, the fault detection result of the detection area is deemed to meet the requirements, and the corresponding detection area is marked as a normal area. Otherwise, the fault detection result of the detection area is deemed to not meet the requirements, and the corresponding detection area is marked as an abnormal area. A fault handling signal is generated and sent to the mobile terminal of the management personnel. If all detection areas are marked as normal areas, the corresponding detection time point is marked as a normal time point. Otherwise, the corresponding detection time point is marked as an abnormal time point.

[0023] The data acquisition value CJie is obtained in real time by sensors deployed within detection area i, while the standard value BZie is pre-stored in a database based on equipment specifications or historical operating data. The detection value JCie is calculated using absolute value operations to eliminate the influence of positive and negative deviations on the detection results. The detection threshold JCed is set according to different parameter types and equipment tolerance; for example, the current parameter threshold is set to ±5% of the rated value, and the voltage parameter threshold is set to ±3%. Detection area i is divided into n independent areas, with m sensors installed in each area to ensure that the data acquisition coverage density meets the detection accuracy requirements.

[0024] Specifically, after the sensor triggers data acquisition at the detection time point, the system automatically retrieves the standard value of the corresponding parameter for comparison. The difference between the acquired value and the standard value is quantified through absolute value to form a comparable detection value. When the detection value exceeds a preset threshold, the device status corresponding to that parameter is determined to be abnormal. By comparing the threshold values ​​of all parameters within the region one by one, the comprehensiveness of anomaly detection is ensured. For example, in detection region i=1, the acquired value CJ11 of parameter e=1 is 215V, the standard value BZ11 is 220V, and the detection value JC11=|215-220|=5V. If the detection threshold JC1d is 6V, the parameter is determined to be normal; if the detection threshold JC1d is 4V, an anomaly determination is triggered. This process replaces manual experience judgment with quantitative calculation, improving the objectivity and response speed of fault detection.

[0025] The detection thresholds for each parameter are pre-stored in a database, with each parameter having its own independent threshold set based on historical operating data or equipment specifications. The detection value is calculated based on the absolute difference between the acquired value and the standard value. For example, if the acquired value of a parameter is 25 and the standard value is 20, the detection value is 5. The detection values ​​of all parameters within the detection area are compared one by one with their corresponding detection thresholds. If the detection value of a parameter exceeds the threshold, an anomaly determination is triggered directly. The generation of fault handling signals uses an automated triggering mechanism. For example, when a detection area is marked as an abnormal area, the system automatically generates a signal containing the abnormal area number and the parameter exceeding the standard, and sends it to the designated terminal through a preset interface.

[0026] Specifically, in practical applications, a testing area for a photovoltaic power station can be set up, for example, an area containing 10 photovoltaic modules. For this testing area, five key parameters (e) are set for testing, including output voltage, output current, surface temperature, tilt angle, and dust coverage. The detection thresholds (JCed) for these five parameters are retrieved from the database, which are: output voltage threshold 300V, output current threshold 8A, surface temperature threshold 60℃, tilt angle threshold 5°, and dust coverage threshold 10%.

[0027] At the detection time point, the system collects the actual detection values ​​JCie of five parameters within the detection area, such as: output voltage 295V, output current 7.8A, surface temperature 58℃, tilt angle 3°, and dust coverage 8%. These detection values ​​are compared with their corresponding detection thresholds, and all values ​​are found to be below their respective thresholds. Therefore, the system determines that the fault detection results for this detection area meet the requirements and marks the detection area as a normal area.

[0028] The operation and maintenance assessment module is used to perform overall operation and maintenance assessment and analysis of the photovoltaic power station. It marks the ratio of the number of abnormal areas to the number of normal areas at each detection time point as the anomaly coefficient; the maximum value of the anomaly coefficient at all detection time points within the detection period as the anomaly performance value; the ratio of the number of times an abnormal time point is marked to the number of times a normal time point is marked within the detection period as the fault performance value; and the sum of the anomaly performance value and the fault performance value as the operation and maintenance requirement value. It retrieves the operation and maintenance requirement threshold from the database and compares the operation and maintenance requirement value with the threshold. If the operation and maintenance requirement value is less than the threshold, the overall operating status of the photovoltaic power station within the detection period is determined to meet the requirements and does not have operation and maintenance characteristics. If the operation and maintenance requirement value is greater than or equal to the threshold, the overall operating status of the photovoltaic power station within the detection period is determined to not meet the requirements and has operation and maintenance characteristics. An operation and maintenance analysis signal is then generated and sent to the operation and maintenance analysis module.

[0029] Among them, the anomaly coefficient dynamically reflects the degree of anomaly at a single detection time point by the ratio of the number of abnormal areas to the number of normal areas, the anomaly performance value captures the most severe abnormal state by selecting the maximum value of the anomaly coefficient within the detection cycle, and the fault performance value quantifies the persistence of the fault by the ratio of the abnormal time point to the normal time point.

[0030] Specifically, within the detection cycle, an anomaly coefficient is calculated at each detection time point based on the number of abnormal and normal areas. The maximum value of the anomaly coefficient characterizes the most severe anomaly within the cycle, while the ratio of abnormal time points to normal time points reflects the frequency of fault occurrence. By adding the anomaly performance value to the fault performance value, the operational requirement value can be obtained, simultaneously considering the impact of anomaly peaks and fault persistence on the overall operational status. For example, if the maximum anomaly coefficient in a certain detection cycle is 0.5, and the ratio of abnormal time points to normal time points is 0.3, then the operational requirement value is 0.8. By comparing it with a preset threshold, it can be determined whether operational maintenance is required. This method improves the objectivity and accuracy of operational assessment by quantifying the degree of anomaly and the frequency of faults, providing a reliable basis for subsequent operational decisions.

[0031] The maintenance requirement thresholds are obtained through training on historical operational data and stored in a preset threshold table in the database. The comparison between the maintenance requirement value and the maintenance requirement threshold uses a numerical comparison algorithm, which is implemented by a computer program to determine the numerical relationship. The generation of maintenance analysis signals is executed by the communication module, and the signal transmission protocol adopts the MQTT IoT communication protocol.

[0032] Specifically, the maintenance requirement threshold serves as the judgment benchmark and is input to the comparison unit after being called through the database interface. The maintenance requirement value is obtained by adding the abnormal performance value and the fault performance value, and is then input to the comparison unit for numerical comparison. When the maintenance requirement value exceeds the threshold, the comparison unit triggers a signal generation command, and the communication module generates an maintenance analysis signal containing a timestamp and the power station number according to the command. For example, if the maintenance requirement threshold is set to 0.8, and the calculated maintenance requirement value within the detection period is 0.85, it is determined that the power station's operating status does not meet the requirements, and an maintenance analysis signal containing the power station ID and the detection period number is generated. This signal is transmitted to the maintenance analysis module via the wireless network, triggering subsequent decision analysis processes to achieve automated processing from status assessment to maintenance decision-making.

[0033] The operation and maintenance analysis module is used to perform operation and maintenance decision analysis for photovoltaic power plants. It marks the parameter e where the detected value JCie at an abnormal time point is not less than the corresponding detection threshold JCed as the analysis object. The ratio of the detected value JCie of the analysis object at the abnormal time point to the detection threshold JCed is marked as the exceeding value of the analysis object. The maximum value of all parameters e corresponding to the exceeding values ​​of the analysis object is marked as the exceeding performance value. The variance of all exceeding performance values ​​of parameter e is calculated to obtain the decision coefficient. The decision threshold is obtained from the database, and the decision coefficient is compared with the decision threshold. If the decision coefficient is less than the decision threshold, a comprehensive maintenance signal is generated and sent to the mobile terminal of the management personnel. If the decision coefficient is greater than or equal to the decision threshold, the L1 parameters e with the largest exceeding performance values ​​are marked as optimization objects, a key optimization signal is generated, and the key optimization signal and the optimization objects are sent to the mobile terminal of the management personnel.

[0034] The decision coefficients, calculated using variance, reflect the differences in exceedance fluctuations among different parameters; a larger variance indicates a more significant difference in the degree of exceedance between parameters. The decision threshold, serving as the critical point for switching maintenance strategies, is obtained through training with historical operation and maintenance data. L1 is a preset positive integer used to control the number of optimization targets, and its value is dynamically adjusted according to the power plant scale. Emphasis is placed on optimizing the parameter identification information carried by the signal, enabling maintenance personnel to quickly locate key parameters.

[0035] Specifically, after calculating the exceedance values ​​of each parameter within the detection period, the dispersion of the exceedance degree of the parameter group is quantified through variance calculation. When the variance value is below a preset threshold, it indicates that the fluctuation of the parameter group's exceedance is small, and a comprehensive maintenance process is triggered to uniformly inspect all parameters. When the variance value exceeds the threshold, it indicates that some parameters have significantly deviated from the standard. At this time, the system automatically extracts the parameters with the highest exceedance values ​​(L1 position) as optimization targets. For example, when a power plant sets L1=3, the system will select the three parameters with the most severe exceedances to generate an optimization list. This mechanism realizes intelligent switching of maintenance strategies through data dispersion analysis, optimizing the allocation of maintenance resources while ensuring system stability.

[0036] The process involves several steps. First, parameters whose detected values ​​at abnormal time points are not less than a detection threshold are marked as analysis objects. In practice, this can be achieved by iterating through all parameter detected values ​​at abnormal time points and selecting those exceeding or equaling the detection threshold. The ratio of the detected value of an analysis object to the detection threshold is marked as the exceeding value. For example, when the detected value is 1.5 times the detection threshold, the exceeding value is 1.5. The maximum value of all exceeding values ​​for each parameter is marked as the exceeding performance value. In practice, this can be achieved by calculating the exceeding values ​​for each parameter at all abnormal time points and taking the maximum value as the exceeding performance value for that parameter. Finally, variance is calculated for all exceeding performance values. For example, the decision coefficients in variance form are obtained by averaging the squares of the deviations of each exceeding performance value from the mean.

[0037] Specifically, within the detection period, parameters whose detection values ​​exceed or equal to the detection threshold at all abnormal time points are first identified and marked as analysis objects. For each analysis object, the ratio of its detection value at the abnormal time point to the corresponding detection threshold is calculated to obtain the exceeding value. The exceeding values ​​of each parameter at all abnormal time points are statistically analyzed, and the maximum value is extracted as the exceeding performance value of that parameter. After collecting the exceeding performance values ​​of all parameters, the variance of these values ​​is calculated, and the variance result is the decision coefficient. Through variance calculation, the dispersion of the exceeding performance values ​​of different parameters can be reflected. The larger the variance, the more obvious the fluctuation difference between parameters. When the decision coefficient is less than the decision threshold, it indicates that the parameter fluctuation difference is small, and comprehensive maintenance is required; when the decision coefficient is greater than or equal to the decision threshold, it indicates that some parameters fluctuate significantly, and the parameter with the largest exceeding performance value should be optimized first. Thus, by analyzing the parameter fluctuation through variance, the targeted nature of the maintenance strategy can be optimized.

[0038] Example 2: Figure 2 As shown, a method for detecting operational faults in a photovoltaic power plant based on data acquisition and analysis includes the following steps: Step 1: Conduct regional testing and analysis of the photovoltaic power station: Generate a testing cycle, set several testing time points with equal time intervals within the testing cycle, and obtain the testing value JCie of parameter e within the testing area i at the testing time points; Step 2: Conduct an overall operation and maintenance assessment and analysis of the photovoltaic power station: obtain the operation and maintenance requirements value for the testing period, and determine whether the overall operating status of the photovoltaic power station meets the requirements within the testing period based on the operation and maintenance requirements value; Step 3: Conduct operation and maintenance decision analysis for photovoltaic power plants: Obtain decision coefficients for the detection cycle, and generate targeted operation and maintenance decision signals based on the decision coefficients.

[0039] A photovoltaic power plant operation fault detection system based on data acquisition and analysis is disclosed. During operation, the regional detection module first divides the photovoltaic power plant into multiple detection zones and deploys sensors in each zone to collect parameters. The system generates a detection cycle, setting multiple equally spaced detection time points within the cycle. At each detection time point, the regional detection module performs fault detection on each zone and marks the detection time point as normal or abnormal. The operation and maintenance assessment module acquires abnormal and fault performance values ​​within the detection cycle, and uses their sum as the operation and maintenance requirement value. By comparing the operation and maintenance requirement value with a preset threshold, it determines whether the overall operating status of the photovoltaic power plant within the detection cycle meets the requirements. The operation and maintenance analysis module performs decision analysis based on the operation and maintenance assessment results and generates corresponding operation and maintenance strategies. The entire system achieves data interaction and information sharing between modules through a database. The regional detection module achieves precise fault location by dividing the power plant into multiple detection zones. Setting the detection cycle and multiple detection time points enables the system to capture dynamically changing fault characteristics. The operation and maintenance assessment module quantifies abnormal and fault performance into operation and maintenance requirement values, providing an objective basis for overall status assessment. The operations and maintenance analysis module formulates targeted operations and maintenance strategies based on the evaluation results, thereby improving operations and maintenance efficiency.

[0040] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

[0041] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0042] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A photovoltaic power plant operation fault detection system based on data acquisition and analysis, characterized in that, It includes a region detection module, an operation and maintenance assessment module, and an operation and maintenance analysis module connected in sequence, and the region detection module, operation and maintenance assessment module, and operation and maintenance analysis module are all communicatively connected to the database; The regional detection module is used to perform regional detection and analysis on the photovoltaic power station: the photovoltaic power station is divided into detection areas i, i=1, 2, ..., n, where n is a positive integer; sensors are deployed according to the equipment in the detection area and parameters e, e=1, 2, ..., m, where m is a positive integer, a detection cycle is generated, and several detection time points with equal time intervals are set within the detection cycle. Fault detection is performed at the detection time points and the detection time points are marked as normal time points or abnormal time points. The operation and maintenance assessment module is used to perform overall operation and maintenance assessment and analysis of the photovoltaic power station: obtain the abnormal performance value and fault performance value of the detection period, mark the sum of the abnormal performance value and fault performance value as the operation and maintenance requirement value, and determine whether the overall operating status of the photovoltaic power station meets the requirements within the detection period through the operation and maintenance requirement value. The operation and maintenance analysis module is used to perform operation and maintenance decision analysis for photovoltaic power plants.

2. The photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 1, characterized in that, The specific process of fault detection at the detection time point includes: collecting the value of parameter e in the detection area i at the detection time point and marking it as the collected value CJie; retrieving the standard value BZie corresponding to parameter e; marking the absolute value of the difference between the collected value CJie and the standard value BZie as the detection value JCie of parameter e; and marking the detection area as a normal area or an abnormal area through the detection value JCie.

3. The photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 2, characterized in that, The specific process of marking a detection area as a normal or abnormal area includes: retrieving the detection threshold JCed of parameter e from the database; comparing the detection values ​​JCie of parameter e within detection area i with the corresponding detection threshold JCed; if all detection values ​​JCie are less than the corresponding detection threshold JCed, the fault detection result of the detection area is determined to meet the requirements, and the corresponding detection area is marked as a normal area; otherwise, the fault detection result of the detection area is determined to not meet the requirements, the corresponding detection area is marked as an abnormal area, a fault processing signal is generated, and the fault processing signal is sent to the mobile terminal of the management personnel.

4. The photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 3, characterized in that, The specific process of marking detection time points as normal or abnormal time points includes: if all detection areas are marked as normal areas, then the corresponding detection time point is marked as a normal time point; otherwise, the corresponding detection time point is marked as an abnormal time point.

5. A photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 4, characterized in that, The process of obtaining abnormal performance values ​​and fault performance values ​​includes: marking the ratio of the number of abnormal areas to the number of normal areas at the detection time point as the abnormal coefficient; marking the maximum value of the abnormal coefficients at all detection time points within the detection period as the abnormal performance value; and marking the ratio of the number of times abnormal time points are marked to the number of times normal time points are marked within the detection period as the fault performance value.

6. A photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 5, characterized in that, The specific process for determining whether the overall operating status of a photovoltaic power station meets the requirements during the testing period includes: retrieving the operation and maintenance requirement threshold from the database and comparing the operation and maintenance requirement value with the operation and maintenance requirement threshold; if the operation and maintenance requirement value is less than the operation and maintenance requirement threshold, it is determined that the overall operating status of the photovoltaic power station meets the requirements during the testing period and does not have operation and maintenance characteristics; if the operation and maintenance requirement value is greater than or equal to the operation and maintenance requirement threshold, it is determined that the overall operating status of the photovoltaic power station does not meet the requirements during the testing period and has operation and maintenance characteristics, generating an operation and maintenance analysis signal and sending the operation and maintenance analysis signal to the operation and maintenance analysis module.

7. A photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 6, characterized in that, The specific process of the operation and maintenance analysis module for performing operation and maintenance decision analysis on photovoltaic power plants includes: obtaining the decision coefficient for the detection cycle, obtaining the decision threshold through the database, and comparing the decision coefficient with the decision threshold; if the decision coefficient is less than the decision threshold, a comprehensive maintenance signal is generated and sent to the mobile terminal of the management personnel; if the decision coefficient is greater than or equal to the decision threshold, the L1 parameters e with the largest out-of-standard performance values ​​are marked as optimization objects, a key optimization signal is generated, and the key optimization signal and the optimization objects are sent to the mobile terminal of the management personnel.

8. A photovoltaic power plant operation fault detection system based on data acquisition and analysis according to claim 7, characterized in that, The process of obtaining the decision coefficients for the detection cycle includes: marking the parameter e where the detection value JCie at the abnormal time point is not less than the corresponding detection threshold JCed as the analysis object; marking the ratio of the detection value JCie of the analysis object at the abnormal time point to the detection threshold JCed as the exceeding value of the analysis object; marking the maximum value of all parameters e as the exceeding performance value of the analysis object; and calculating the variance of all exceeding performance values ​​of parameters e to obtain the decision coefficients.

Images