Cloud computing based photovoltaic performance evaluation system

Abstract

The present application relates to the technical field of photovoltaic performance evaluation, in particular to a photovoltaic performance evaluation system based on cloud computing, which comprises an electrical thermal response interpretation module, an abnormal aggregation distribution module, a rhythm deviation trajectory module, a voltage cumulative deviation module and a risk early warning judgment level module. Through the cooperative comparison of thermal response ratio and electrical signal, the present application fuses thermal and electrical characteristics with voltage changes in multi-dimensional time series, greatly improves the linkage abnormality recognition ability of heterogeneous parameters, focuses on spatial structure distribution and number mapping, realizes hierarchical division of abnormal aggregation phenomenon, strengthens dynamic risk insight of spatial dense area, constructs dynamic deviation mode based on multi-day minute-level operation data, accurately captures implicit abnormalities such as continuous deviation and reverse turning in historical and real-time change trends, and refines and archives voltage trends and cumulative characteristics, thereby laying a foundation for multi-node and multi-period risk positioning through classified summary, and realizing multi-source synchronous abnormal segment alignment processing mode.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic performance evaluation, and particularly relates to a photovoltaic performance evaluation system based on cloud computing. BACKGROUND

[0002] Photovoltaic performance evaluation mainly involves comprehensive monitoring, analysis and evaluation of the running state, power output, conversion efficiency and environmental adaptability of photovoltaic power generation systems. This field combines sensor technology, data acquisition systems, meteorological models, electrical modeling and data analysis algorithms to accurately evaluate the actual performance of photovoltaic components and systems, discover potential faults, improve power generation efficiency, and provide the basis for system optimization and operation and maintenance decisions.

[0003] Among them, the photovoltaic performance evaluation system based on cloud computing is a system that uses a cloud computing platform to remotely collect, centrally process and intelligently analyze photovoltaic power generation system operation data. It obtains photovoltaic component, grid interface and environmental parameter data through distributed data acquisition terminals and uploads them to the cloud for real-time processing and comprehensive evaluation. Its purpose is to improve the intelligence and automation level of photovoltaic system performance evaluation, support remote monitoring, fault diagnosis, operation optimization and operation and maintenance decisions, and is widely used in photovoltaic power stations, distributed photovoltaic systems and intelligent energy management platforms.

[0004] The existing technology generally relies on single signal independent monitoring and dispersed data acquisition, lacks depth analysis of parameter fusion and spatial correlation, and leads to the inability to identify multiple abnormal events in heterogeneous data, difficulty in tracking and positioning spatial risk points, long-term dependence on fixed periodic collection and one-way trend judgment, difficulty in timely reflecting continuous deviation and trend changes, easy covering of weak abnormalities in history and real-time changes, lack of classification and arrangement of voltage data accumulation characteristics, inability to reveal risk propagation paths among multiple nodes, lack of collaborative alignment of multi-source data, risk identification in complex environments is prone to omission or misjudgment, performance evaluation is limited to rough judgment and single-point alarm stage, and the reliability of intelligent management and proactive operation is affected. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art, and a photovoltaic performance evaluation system based on cloud computing is proposed.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: a photovoltaic performance evaluation system based on cloud computing, the system comprises:

[0007] The electrical thermal response interpretation module is based on a cloud computing platform, analyzes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic component, calculates the ratio and compares it with the same period of the previous day, identifies the ratio change trend, selects the key time period, performs time alignment, and obtains synchronous thermal and electrical abnormal characteristics;

[0008] The abnormal aggregation distribution module analyzes the abnormal segment distribution of the components in the region based on the synchronous thermoelectric abnormality feature, calculates the cumulative number of abnormal segments, judges the aggregation degree, screens the key number interval, optimizes the correspondence between the number and the spatial coordinates, and generates the regional abnormal aggregation distribution feature;

[0009] The rhythm deviation trajectory module analyzes the current and historical current sequence of the photovoltaic component based on the regional abnormal aggregation distribution feature, calculates the current deviation trajectory, judges the continuous deviation section and its direction change, combines the component number marking position, and outputs the current deviation dynamic feature.

[0010] The voltage cumulative deviation module analyzes the voltage data of each node based on the current deviation dynamic feature, calculates the voltage change trend, screens the voltage segment with continuous deviation and cumulative change characteristics, judges whether the cumulative change is abnormal, and obtains the voltage cumulative deviation feature.

[0011] The present application improves that the synchronous thermoelectric abnormality feature includes abnormal response ratio type, associated voltage change feature, key time label, the regional abnormal aggregation distribution feature includes aggregation distribution type, spatial density, number group identification result, the current deviation dynamic feature includes continuous deviation category, change direction attribute, number mapping information, and the voltage cumulative deviation feature includes cumulative amplitude attribute, classification result, node response group.

[0012] The present application improves that the electrical thermal response interpretation module includes:

[0013] The ratio sequence calculation submodule analyzes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic component based on the cloud computing platform, calculates the ratio of current drop amplitude and temperature rise speed at each time point, compares the ratio change of consecutive time points, and obtains the ratio sequence trend.

[0014] The key time period screening submodule judges the change trend of the ratio sequence in each time period according to the ratio sequence of the same period of the previous day, screens the time period with key change of the ratio, and obtains the ratio change key interval.

[0015] The thermoelectric abnormality feature extraction submodule analyzes the voltage change of the corresponding time period based on the ratio change key interval, judges the synchronous feature of the ratio change and the voltage change, performs time alignment, and obtains the synchronous thermoelectric abnormality feature.

[0016] The present application improves that the abnormal aggregation distribution module includes:

[0017] The number space matching submodule analyzes the correspondence between the number of the photovoltaic module and the spatial position based on the synchronous thermoelectric anomaly feature, identifies the number of each photovoltaic module, performs data pairing of the number and the spatial coordinate, and obtains the number space correspondence;

[0018] The anomaly accumulation calculation submodule calculates the total amount of anomaly segments of each number in the spatial range based on the number space correspondence, analyzes the spatial distribution of the anomaly segment accumulation number, compares the accumulation difference under each number, judges the distribution trend of the anomaly segments of the same region number, and obtains the total number of anomaly segment accumulations;

[0019] The aggregation interval screening submodule screens the accumulation number in each number interval according to the total number of anomaly segment accumulations, compares the spatial distribution data of the number interval, calculates the continuous distribution of the anomaly segments of the number interval, calculates the anomaly aggregation degree of the number interval, and generates the regional anomaly aggregation distribution feature.

[0020] The present application improves that the rhythm deviation trajectory module comprises:

[0021] The current sequence acquisition submodule analyzes the collected current data of the photovoltaic module every minute for consecutive days based on the regional anomaly aggregation distribution feature, arranges the daily current data in time sequence in combination with the module number index, compares the data integrity and continuity through daily classification and archiving operation, and obtains a current sequence set;

[0022] The deviation section identification submodule calls the current sequence set, calculates the absolute difference between the current of each minute in the current running day and the current of each minute in the same period in the history, screens the minute section of the daily continuous deviation, judges the continuous deviation and direction change, calculates the deviation section feature value, and obtains the current deviation section;

[0023] The current deviation feature output submodule analyzes the section range of the current deviation section, optimizes the module number index, compares the section feature and the actual position of the module, integrates each type of data, and outputs the current deviation dynamic feature.

[0024] The present application improves that the voltage accumulation deviation module comprises:

[0025] The voltage trend calculation submodule analyzes the voltage data continuously collected by each node based on the current deviation dynamic feature, calculates the voltage change amplitude of each node at adjacent collection time, judges the continuous change of the voltage at each collection stage, analyzes the voltage change accumulation trend of each stage, and obtains the voltage change trend;

[0026] The continuous offset screening submodule judges whether the voltage increasing / decreasing direction of each acquisition stage in the voltage change trend is consistent continuously, analyzes the coverage interval and direction continuity of the continuous offset stage, compares the offset direction characteristics between the difference segments, screens the typical segments of the voltage continuous offset, and obtains the continuous offset segment;

[0027] The segment anomaly judgment submodule analyzes the voltage change interval of the continuous offset segment, calculates the continuity of the voltage change direction and the fluctuation distribution in each segment, judges whether the change performance deviates from the normal state, and obtains the voltage cumulative offset characteristics.

[0028] The system further comprises:

[0029] The risk early warning judgment module matches the current offset dynamic characteristics and synchronous thermoelectric anomaly characteristics in the same period based on the voltage cumulative offset characteristics, judges the synchronous distribution of the abnormal segment in time and component number, analyzes the monitoring log and combines the climate fluctuation data, and obtains an array risk grading index;

[0030] The array risk grading index comprises a risk level category, a risk interval identifier, and an associated influencing factor.

[0031] The risk early warning judgment module comprises:

[0032] The voltage offset detection submodule analyzes the voltage change curve of the photovoltaic component in the same monitoring period based on the voltage cumulative offset characteristics, judges the difference of the voltage fluctuation trend between each component, screens the component number with abnormal change trend, and generates a voltage trend distribution;

[0033] The current thermoelectric synchronous judgment submodule compares the current fluctuation characteristics and the thermoelectric anomaly signal of the voltage trend distribution corresponding period, judges the synchronism of the current and the thermoelectric signal, screens the data segment with synchronous anomaly, and generates a synchronous anomaly distribution;

[0034] The array risk grading submodule calculates the abnormal distribution of the components involved in the synchronous anomaly distribution in the monitoring log and the climate data fluctuation interval, analyzes the synchronous distribution range and the abnormal data characteristics of each component, optimizes the data corresponding relationship, and obtains an array risk grading index.

[0035] Compared with the prior art, the application has the advantages and positive effects that:

[0036] In the present application, by the synergistic comparison of the thermal response ratio and the electrical signal, the thermal and electrical characteristics and the voltage change are fused in a multi-dimensional time sequence, which greatly improves the linkage abnormality recognition ability of heterogeneous parameters, focuses on the spatial structure distribution and the number mapping, realizes the hierarchical division of abnormal aggregation phenomenon, strengthens the dynamic risk insight of spatial dense area, constructs the dynamic deviation mode across multi-day minute-level operation data, accurately captures the implicit abnormalities such as continuous deviation and reverse turning in the historical and real-time change trend, and refines and archives the voltage trend and cumulative characteristics, which lays a foundation for multi-node and multi-period risk positioning through classified summary, multi-source synchronous abnormal segment alignment processing mode, global risk grading under the interaction of time, number and environmental factors, and improves the fine early warning level of large-scale photovoltaic system. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 is a system flowchart of the present application;

[0038] Figure 2 is a flowchart of the electrical thermal response interpretation module in the present application;

[0039] Figure 3 is a flowchart of the abnormal aggregation distribution module in the present application;

[0040] Figure 4 is a flowchart of the rhythm deviation trajectory module in the present application;

[0041] Figure 5 is a flowchart of the voltage cumulative deviation module in the present application;

[0042] Figure 6 is a flowchart of the risk early warning judgment module in the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0044] In the description of the present application, it should be understood that the orientations or positional relationships indicated by the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0045] EMBODIMENT

[0046] Referring to Figure 1 The application provides a technical solution: a photovoltaic performance evaluation system based on cloud computing includes:

[0047] The electrical thermal response interpretation module is based on a cloud computing platform, analyzes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic module, calculates the ratio of the two at each time point, compares the current ratio sequence with the ratio sequence at the same period of the previous day, judges the change trend of the ratio in multiple periods, filters the time period critical to the change of the ratio, synchronously analyzes the voltage change information, performs time alignment, and obtains the synchronous thermal-electric abnormality feature;

[0048] The abnormal aggregation distribution module is based on the synchronous thermal-electric abnormality feature, calls the photovoltaic module number and spatial position, analyzes the abnormal segment distribution of the components in the same area, calculates the cumulative number of abnormal segments under each number, judges the aggregation degree of the abnormal segments in the number interval, filters the number interval critical to the aggregation phenomenon, optimizes the correspondence between the number and the spatial coordinates, and generates the regional abnormal aggregation distribution feature;

[0049] The rhythm deviation trajectory module is based on the regional abnormal aggregation distribution feature, analyzes the current output sequence of the photovoltaic module every minute for multiple consecutive days, compares the current minute current with the historical current sequence of the same period using the cloud platform, calculates the current deviation change trajectory, judges the continuous deviation section and its direction change, combines the component number index to mark the position, and outputs the current dynamic deviation feature;

[0050] The voltage cumulative deviation module is based on the current dynamic deviation feature, analyzes the voltage data of each node, calculates the voltage change trend, filters the voltage segment with cumulative change characteristics that continuously deviates, judges whether the cumulative change of the segment is abnormal, adjusts the segment classification and archiving method, and obtains the voltage cumulative deviation feature;

[0051] The risk early warning judgment module is based on the voltage cumulative deviation feature, matches the current dynamic deviation feature and the synchronous thermal-electric abnormality feature in the same period, judges the synchronous distribution of the abnormal segments in time and component number, analyzes the monitoring log of the photovoltaic module on the cloud platform, and judges the influence degree in combination with the climate fluctuation data to obtain the array risk grading index.

[0052] The synchronous thermal-electric abnormality feature includes abnormal response ratio types, associated voltage change characteristics, and key time labels, the regional abnormal aggregation distribution feature includes aggregation distribution types, spatial density, and number group identification results, the current dynamic deviation feature includes continuous deviation categories, change direction attributes, and number mapping information, the voltage cumulative deviation feature includes cumulative amplitude attributes, classification results, and node response groups, and the array risk grading index includes risk level categories, risk interval identifiers, and associated influencing factors.

[0053] The ratio sequence refers to the sequence of the ratio between the current drop amplitude and the temperature rise speed arranged in time sequence; the ratio change key refers to the key period in which a significant change or mutation occurs in the ratio sequence. The component number refers to the unique identity code of each photovoltaic component in the system; the abnormal segment refers to the time period in which the operation data is detected to be abnormal; the aggregation degree refers to the intensity of the concentrated distribution of multiple abnormal segments in space or number; the aggregation phenomenon key refers to the key number interval in which the abnormality appears highly concentrated in the abnormal segment distribution. The current output sequence refers to the actual recorded output current data sequence of the photovoltaic component in continuous days; the current deviation change track refers to the deviation change process record formed after comparing the current with the historical current data of the same period; the direction change situation refers to the change turning point of the deviation track in the time sequence, which changes from increasing to decreasing or from decreasing to increasing. Each node refers to the monitoring point of different data collection or electrical connection in the photovoltaic system; the cumulative change feature refers to the performance feature of the voltage data continuously accumulated in a certain period of time; the voltage segment refers to the voltage monitoring data time period with a specific deviation trend after being identified; the classification and archiving method refers to the method of classifying and archiving the voltage abnormal segment according to the segment characteristics. The influence degree refers to the influence size or severity of the abnormal distribution of the array on the overall performance and risk level of the photovoltaic component.

[0054] Please refer to Figure 2 , the electrical thermal response interpretation module comprises:

[0055] The ratio sequence calculation submodule is based on a cloud computing platform, analyzes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic component, calculates the ratio of the current drop amplitude and the temperature rise speed at each time point, compares the ratio change of the continuous time points, and obtains the ratio sequence trend;

[0056] The cloud computing platform processes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic module. First, the specified photovoltaic module current and temperature monitoring data during operation are extracted from the database, and each set of data is traversed in time sequence. The current change amplitude of each two consecutive sampling time points is determined, the drop amplitude of the time point is obtained by directly subtracting the current values of adjacent time points, and the current change amplitude of the entire sequence is completed in turn. The temperature data in the same time period is processed in the same way to obtain the temperature rise speed sequence. The current change amplitude of each time point is paired with the temperature rise speed at the same time point, the ratio of the two is calculated, and the ratio data of all time points are gradually filled to form a complete ratio sequence. In practical application, the current monitoring data of a photovoltaic module from 8:00 to 8:04 is 5.0, 4.8, 4.6, 4.3, and 4.0, and the temperature monitoring data is 30, 31, 33, 35, and 37. First, the current drop amplitude of each minute is calculated as 0.2, 0.2, 0.3, and 0.3, and the temperature rise speed is 1, 2, 2, and 2. Then, the ratio of each current change and temperature rise speed is calculated to obtain the ratio sequence as 0.2, 0.1, 0.15, and 0.15. Further, the ratio data is compared according to the time point, the ratio of the adjacent two time points is compared, and it is determined whether the ratio change exceeds a certain value, such as greater than 0.05, which is marked as a significant change. All time points with changes are marked separately to obtain the ratio change trend sequence, and the ratio sequence trend is obtained as the basis for subsequent analysis.

[0057] The key time period screening submodule determines the change trend of the ratio sequence in each time period according to the ratio sequence trend and the ratio sequence of the same time period of the previous day, and screens the time period with a key change in the ratio to obtain the ratio change key interval.

[0058] The ratio sequences of the same time period of the current day and the previous day are compared one by one. The data of the corresponding time period of the previous day of the same component is called from the cloud platform historical data, the ratio sequences of the current day and the previous day are extracted, and the time points are paired and compared. For the ratio of each time point, the difference between the two is directly calculated. For the time points with a difference greater than 0.05, it is determined that a key change has occurred. All time points determined to have a key change and their surrounding time periods are classified as a ratio change key interval. Assuming that in actual operation, the current sequence is 0.2, 0.1, 0.15, and 0.15, and the previous day sequence is 0.18, 0.1, 0.12, and 0.13, the difference between the two sequences is 0.02, 0, 0.03, and 0.02. No point exceeds 0.05. If the third point of the current sequence becomes 0.25, the difference is 0.02, 0, 0.13, and 0.02. The third point is classified as a key change, and the third time point and its surrounding time points are classified into a key interval. All time periods with significant changes in the ratio are collected in time sequence to form a ratio change key interval.

[0059] The thermoelectric anomaly feature extraction submodule analyzes the voltage change in the corresponding time period based on the ratio change key interval, judges the synchronization feature of the ratio change and the voltage change, performs time alignment, and obtains the synchronized thermoelectric anomaly feature;

[0060] According to the obtained ratio change key interval, the voltage data needs to be further processed. The voltage monitoring data of the same component in the time period corresponding to the ratio change key interval is extracted. The voltage data is matched at each time point in the key interval in turn. The voltage change value is searched according to the time point. The trend of the voltage change is compared with the trend of the ratio change. The time points at which the voltage change amplitude is greater than 0.5 and the ratio change amplitude occurs simultaneously are identified as synchronous changes. For example, the voltage data in the key interval is 38.0, 37.2, 36.9, and 36.5. The ratio data is 0.2, 0.25, 0.15, and 0.15. The change amplitudes of the voltage and the ratio at each point are judged. The first point voltage change is 0.8, and the ratio change is 0.05. The synchronous change is not synchronized. The second point voltage change is 0.3, and the ratio change is 0.1. After screening, all the points of the synchronous change are recorded. The time point, the ratio change, and the voltage change are combined to mark all the synchronous anomaly segments. The ratio, the associated voltage data, and the time point in the segment are summarized as the synchronized thermoelectric anomaly feature.

[0061] Please refer to Figure 3 The anomaly aggregation distribution module includes:

[0062] The number space matching submodule analyzes the corresponding relationship between the photovoltaic module number and the spatial position based on the synchronized thermoelectric anomaly feature, identifies the number of each photovoltaic module, performs data pairing of the number and the spatial coordinates, and obtains the number space correspondence.

[0063] The unique number of each photovoltaic module associated with the thermoelectric anomaly record is retrieved, and the number of each anomaly feature corresponding to the module is retrieved in the data system. All anomaly features are associated with their corresponding numbers, and the spatial coordinate information corresponding to the number is searched in the spatial layout database of the photovoltaic power station field area. The system locates the actual geographic coordinates of the module according to the number in the field area arrangement table. The number and spatial coordinate are paired one by one, and the number and its specific spatial position data form a unique mapping relationship. According to the mapping table, all uploaded anomaly features are sorted and archived according to the number and spatial coordinate. In the actual scene, the numbers A101, A102, A103, and A104 correspond to the coordinates (10, 20), (10, 21), (11, 20), and (11, 21), respectively. If A101 and A103 are recorded as synchronous thermoelectric anomalies at the same time, the numbers and coordinates of A101 and A103 are bound and stored with the anomaly event, and the numbers are sorted in the data list to ensure that each anomaly record has a number and a spatial position. In subsequent operations, the number can be directly located to the spatial coordinate to complete the data pairing of the number and the spatial coordinate, and the number-space correspondence relationship is obtained.

[0064] The anomaly accumulation calculation submodule calculates the total number of anomaly segments of each number in the spatial range based on the number-space correspondence relationship, analyzes the spatial distribution of the anomaly segment accumulation number, compares the accumulation difference under each number, and judges the distribution trend of the anomaly segment of the same regional number to obtain the total number of anomaly segment accumulations.

[0065] All identified photovoltaic module numbers in the spatial region are traversed, the number of anomaly segments of each number is counted in the anomaly feature archive library, and the accumulation count parameter of each number is set. The system retrieves the occurrence time of each number corresponding anomaly feature, adds the number of anomaly segments of the same number appearing at different time periods, forms the anomaly accumulation value of each number, and arranges the accumulation values of the numbers in each spatial unit during the anomaly accumulation statistics. The numbers with a difference greater than 3 times in the same spatial unit are marked as regions with a distribution trend deviation. In the example, the spatial unit numbers A101-A110, if A101 has 5 anomaly segment accumulations, A102 has 2, A103 has 6, and A104 has 1, the system marks the numbers with 5 and 6 accumulations as high frequency, and the numbers with 2 and 1 accumulations as low frequency. The high frequency numbers are further checked for continuous distribution in the spatial coordinate. If A101 and A103 are in the same row, the spatial coordinate corresponding to the row is recorded as an anomaly concentration area. Finally, the total number of anomaly segment accumulations in all spatial units is counted, and the number interval with an accumulation number higher than 1.5 times the average value is marked as an anomaly aggregation, and the total number of anomaly segment accumulations is output.

[0066] The screening submodule of the aggregation interval filters the cumulative number in each numbered interval according to the cumulative total number of abnormal fragments, compares the spatial distribution data of the numbered interval, calculates the continuous distribution condition of the abnormal fragments of the numbered interval, and uses the formula:

[0067] ;

[0068] Calculate the abnormal aggregation degree of the numbered interval The abnormal aggregation degree of the numbered interval is an index for evaluating whether the abnormal phenomenon of the photovoltaic module in a certain interval is dense or concentrated, reflecting the spatial characteristics of local risk aggregation, generating regional abnormal aggregation distribution characteristics, wherein, represents the cumulative number of abnormal fragments of the photovoltaic module in the numbered interval, represents the number of photovoltaic modules in the numbered interval, represents the spatial position distribution difference corresponding to the numbered interval, represents the maximum number of consecutive occurrences of abnormal fragments in the numbered interval, represents the number of modules with abnormal fragments in the numbered interval, represents the total number of photovoltaic modules in the numbered interval. Filter the cumulative number in each numbered interval, group all module numbers according to intervals, each group containing 10 numbers, for example, the first group numbers are 001 to 010, the second group is 011 to 020, and so on, analyze the cumulative number of abnormal fragments in each interval, first calculate the sum of the cumulative number of abnormal fragments of all modules under each interval number, then calculate the total number of modules in the interval, and then analyze the spatial coordinate data of all modules in the interval, use the mean and variance of spatial coordinates to measure the distribution, select the spatial variance as the reference of spatial distribution difference, further in each numbered interval, check whether the number of consecutive modules with abnormal fragments is greater than zero, take the consecutive number greater than zero as the consecutive abnormal fragments, count the maximum number of consecutive abnormal fragments in the interval, and at the same time count the number of numbered intervals with all abnormal fragment cumulative numbers greater than zero, put the data into the formula, take the numbered interval 001-010 as an example, assuming that the cumulative number of abnormal fragments under the numbered interval is 4, 2, 5, 3, 1, 3, 2, 0, 1, and 2, respectively, then the total number of abnormal fragments in the interval is: ;

[0069]

[0070] ;

[0071] ​​​​​​Total number of components in the number interval , spatial variance The maximum number of continuous abnormal fragments (for example, by spatial coordinate statistics) (for example, numbers 003, 004, 005 have continuous abnormal fragments), the number of abnormal components (that is, the number of numbers with cumulative number greater than zero is 8), put into the formula:

[0072] ;

[0073] If the obtained value is greater than 1.5, it means that the abnormal fragment aggregation degree in the number interval is high, which can be used as the key area for subsequent spatial optimization and abnormal aggregation area screening. The regional abnormal aggregation distribution characteristics are calculated, and the result is 1.946, which represents that there is a certain abnormal fragment aggregation trend in the number interval.

[0074] Please refer to Figure 4 , the rhythm deviation trajectory module includes:

[0075] The current sequence acquisition submodule analyzes the collected photovoltaic component current data every minute for multiple days based on the regional abnormal aggregation distribution characteristics, combines the component number index, arranges the daily current data in time sequence, compares the data integrity and continuity through daily classification and archiving operation, and obtains the current sequence set;

[0076] The system retrieves all photovoltaic module numbers within the corresponding spatial distribution area, and then filters each module number one by one. It extracts the minute-by-minute current measurement values ​​for that module number over the most recent three consecutive days from historical operational data records. The system indexes and binds the extracted current data with the module number on a daily basis. For example, assuming module numbers are B201, B202, and B203, the corresponding current monitoring data from 8:00 to 8:05 each day for the three days are: B201: 4.6, 4.7, 4.8, 4.8, 4.7, 4.6; B202: 5.1, 5.2, 5.3, 5.3, 5.2, 5.1; B203: 4.2, 4.2, 4.3, 4.3, 4.2, 4.1. The system then iterates through the original database in ascending chronological order for each sampling time, matching all current values ​​for the same module number on the same day. The current sampling points are arranged in chronological order, and the system automatically archives them daily. The daily data is stored separately and then the integrity of the daily current data within three days is checked by comparing whether the number of sampling points per day is consistent with the number of data points that should be collected within the theoretical period. If more than 1% of the data points are missing on a certain day, the data for that day is marked as incomplete; otherwise, it is recorded as complete. All complete and continuous daily sequences are classified and stored in the current sequence set. During the data processing, the continuity and integrity of the current data with the same number are judged by comparing the daily data sequences of each number within three consecutive days. If B201 has fewer than two sampling points on a certain day, the archiving for that day is abnormal. If the data of B202 and B203 are collected completely every day, they are all included in the sequence set, resulting in a current sequence set indexed by number, archived daily, and with complete data.

[0077] The deviation segment identification submodule calls the current sequence set to calculate the absolute difference between the current per minute of the current operating day and the current per minute of the same period in history, and filters out the minute segments with continuous deviations each day using the formula:

[0078] ;

[0079] Determine continuous deviations and changes in direction, and calculate the characteristic values ​​of the deviation segments. The current deviation range is obtained, representing the overall degree of deviation of the photovoltaic module's current change during the operating day from its historical performance. No. Current output data for the current operating day (minutes). It is the first Historical current output data for the same period in minutes. This is the total number of minutes in the current running day. It is the first Outlier current characteristic parameters of anomalous clustering regions This represents the number of abnormal clusters within the area to which the current component belongs. It is the first a fluctuation range or a dispersion degree) of the historical current data of the same period of the minute history;

[0080] The absolute difference between the current and the historical current of the same period is calculated for each minute of the current operating day, for example, the currents collected by the A1 component from 9:00 to 9:02 on May 12, 2025 are 3.0, 3.2, and 3.1 amperes, respectively, and the historical current of the same period is 3.1, 3.3, and 3.0 amperes, respectively, corresponding to absolute differences of 0.1, 0.1, and 0.1 amperes, respectively. amperes, amperes, amperes, when screening the minute section of the daily continuous deviation, if the absolute difference is not less than 0.05 amperes for three consecutive minutes, the period is included in the deviation section, the abnormal aggregation points of the area where the A1 component is located are counted, and it is assumed that there are two abnormal aggregation points with out-of-gauge current characteristic parameters of 2.0 amperes and 1.5 amperes, respectively, the sum of the two is 3.5 amperes, and the square root is amperes, a total of 1440 minute data is collected in the current day, and the total historical current fluctuation range is assumed to be 0.04 amperes for each minute, and the cumulative value is 57.6 amperes, and the denominator is The above parameters are substituted into the formula:

[0081] ;

[0082] The section GQ is 0.00145, and if the judgment reference interval is set to 0.002-0.003, the section is judged to be not significantly deviated, the abnormal aggregation points and the historical fluctuation are involved in the deviation characteristic judgment, and the current deviation section is output.

[0083] The current deviation characteristic output sub-module analyzes the section range of the current deviation section, optimizes the component number index, compares the section characteristics with the actual position of the component, and integrates each type of data to output the current deviation dynamic characteristics;

[0084] The section range of the current deviation section is analyzed, if the A1 component is continuously higher than 0.002 in the time period from 9:00 to 10:00 on May 12, 2025, this hour is determined as the deviation section, the component number index is optimized, all the minute current data involved in the time period from 9:00 to 10:00 is corresponded to the A1 component number, the section characteristics are compared with the actual physical position of the A1 component, the GQ of all minutes in the section, the absolute difference of each minute current, and the fluctuation range are counted, the average and the maximum of the GQ in the time period from 9:00 to 10:00 are taken, for example, the average GQ of the section is 0.0025, and the maximum GQ is 0.003, the historical standard deviation and the abnormal aggregation point parameters in the section are summarized, and all types of data are integrated to form the section dynamic characteristic data, including the start and end time of the section, the associated component number, the average and maximum GQ, the absolute difference sequence of the minute current, etc., and the current deviation dynamic characteristics are output.

[0085] Referring to Figure 5 , the voltage cumulative deviation module comprises:

[0086] The voltage trend calculation submodule analyzes the voltage data continuously collected at each node based on the current deviation dynamic characteristics, calculates the voltage change amplitude of each node at adjacent collection time points, judges the continuous change of the voltage at each collection stage, analyzes the cumulative trend of the voltage change at each stage, and obtains the voltage change trend.

[0087] The continuous voltage collection sequence of each node is called, the voltage sampling data corresponding to each node is arranged in chronological order into a complete sequence, the voltage values of adjacent time points are directly subtracted for each node, the voltage change amplitude between each two time points is obtained, the voltage change amplitudes in all time periods are sequentially counted and archived according to the collection time, the system collects all the voltage change amplitudes of each node in the whole cycle, judges whether the voltage of each period increases or decreases, compares the voltage values of the current time and the previous time directly, if the current value is greater than the previous value, it is determined that the voltage increases, otherwise, it is determined that the voltage decreases, the positive and negative directions of the voltage change of each period are recorded in the whole cycle data, and the continuous change intervals of each stage are arranged and merged, for the same node, if the voltage of five or more consecutive periods is positively changed, it is classified into the rising trend stage, and if the voltage of five or more consecutive periods is negatively changed, it is classified into the falling trend stage, the cumulative change amount of the voltage change of each continuous stage is summed and counted, and the result is archived with the stage duration, node number and start and end time, in actual application, for example, the collected data of a node from 8:00 to 8:10 is 37.1, 37.3, 37.6, 37.8, 38.0, 37.7, 37.5, 37.2, 37.0, 36.8 and 36.5, the system sequentially calculates the voltage change values of each time period as 0.2, 0.3, 0.2, 0.2, -0.3, -0.2, -0.3, -0.2, -0.2, -0.3, the cumulative value of the continuous positive segment is 0.2+0.3+0.2+0.2=0.9, the duration is five minutes, and the rising trend is classified, the cumulative value of the subsequent negative segment is -0.3-0.2-0.3-0.2-0.2-0.3=-1.3, and the falling trend segment is classified, after the segmented cumulative counting of all data is completed, the voltage change trend of each node is obtained.

[0088] The continuous deviation screening submodule judges whether the voltage increase and decrease directions of each collection stage in the voltage change trend are consistent, analyzes the coverage interval and direction continuity of the continuous deviation stage, compares the deviation direction characteristics between the difference segments, screens the typical segments of the continuous voltage deviation, and obtains the continuous deviation segment.

[0089] According to the segmented trend, the voltage change direction of each node in each collection stage is manually labeled, and the direction attribute of each stage is compared. If the voltage change of all time points in a stage is positive, the direction of the stage is defined as positive, if all are negative, it is negative, if the direction attribute appears switching in a stage, the switching point is taken as the boundary to further segment, the duration and coverage interval of each stage are counted, the segments with stage length more than five minutes and no switching of direction attribute are divided into continuous offset stages, and the adjacent two stages are compared to analyze the change characteristics of the direction between stages, such as stage one is positive and stage two is negative, it is judged that the direction switches, if the directions of two stages are the same, it is considered as continuous direction, actually, if the voltage change direction of node collection segment A is positive, lasting eight minutes, and the voltage change direction of collection segment B is negative, lasting six minutes, A and B are two independent continuous offset segments, if the direction of segment C is consistent with B and continuous, it is merged into a longer continuous offset segment, after the direction attribute division and comparison of all collected data are completed, the continuous offset segments are screened, the segments with time length more than five minutes, no switching of direction and cumulative change greater than 0.5 are extracted, finally, the segment number, node number, time period start and end point, direction attribute are taken as labels for archiving, and the continuous offset segments are obtained.

[0090] Segment anomaly judgment submodule analyzes the voltage change interval of continuous offset segment, calculates the continuity and fluctuation distribution of voltage change direction in each segment, judges whether the change performance deviates from the normal state, uses the formula:

[0091] ;

[0092] Get voltage cumulative offset characteristics , which is used to quantify the offset performance of voltage segment in continuous time period, wherein, represents the voltage data of the th time point, which is the voltage observation value recorded by the node at the time point, represents the voltage data of the th time point, which is used for time difference comparison with , represents the direction consistency discriminant factor of the th segment, which is used to indicate the consistency degree of voltage change direction in the time period, and takes positive and negative signs for direction weighting, represents the interval fluctuation width corresponding to the voltage change of the th segment, which reflects the distribution range of the voltage in the change process, represents the time span identifier of the th voltage segment, which reflects the duration length of the corresponding voltage change in time, represents the total number of time periods in the segment;

[0093] All voltage observation data within the selected continuous migration segment were arranged in chronological order, and the changes were calculated for each adjacent time point. To ensure comparability of different physical quantities in the feature calculations, [the following steps were taken]. and Normalization was performed. Specifically, the minimum-maximum normalization method can be used. For example, assuming that the voltage data collected by node A at 10:00, 10:05, and 10:10 are 230V, 232V, and 235V respectively, then the first change is... V, the second change is V, and simultaneously determine the consistency of the direction of each voltage change. If both changes are positive, then the corresponding direction consistency discrimination factor is... , When processing the fluctuating distribution, the following steps are taken respectively: and Analyzing the voltage fluctuation range, if we directly use the range, then... V, V, corresponding to time spans such as 10:00-10:05 and 10:05-10:10, are both 5 minutes and can be assigned respectively. , ,at this time , , Substitute the data into the formula to perform the calculation:

[0094] ;

[0095] If similar historical fragments are under normal circumstances Mostly distributed in The interval, then this segment The value is higher than the normal range, and it is judged as an abnormal offset segment. The calculation logic is to couple the voltage changes and direction consistency factors of each time period with the weighted accumulation and the joint normalization of fluctuation width and time span, and integrate the multi-dimensional segment characteristics into a set of values. The results show that 0.63 reflects that the voltage accumulation shift of the segment is obvious and belongs to the abnormal performance. It can be further used as a voltage accumulation shift feature for global analysis.

[0096] Please see Figure 6 The risk warning and classification module includes:

[0097] The voltage offset detection submodule analyzes the voltage change curve of photovoltaic modules within the same monitoring period based on the voltage cumulative offset characteristics, determines the differences in voltage fluctuation trends among modules, filters out module numbers with abnormal change trends, and generates voltage trend distribution.

[0098] The voltage change curve of each photovoltaic module in the respective monitoring period is called, the voltage collection data of each module in the same monitoring period is arranged according to the number, the voltage curve data of each module number in the corresponding time period is extracted, the voltage change curves of all modules in the same period are compared in turn, the voltage value change amplitude of each module in the same time period is calculated, the overall rising, falling or fluctuation characteristics of the voltage curve are judged, the voltage change difference between different modules at the same time or in the same time interval is compared, and the voltage change amplitude of the module and the average change value of the module in the same period are taken as the reference in the comparison process. If the voltage change amplitude of a certain module is greater than 1.2 times of the average value of the same group, it is determined that the change trend is abnormal, otherwise, if it is less than 0.8 times of the average value, it is determined that the trend is lower than the normal state. In the example, if the voltage of A module decreases by 2.0V in 10 minutes, and the average decrease of the group is 1.3V, then A number is determined to be abnormal. The numbers of the modules with abnormal voltage change trend are archived, the numbers of the modules with abnormal voltage change trend are selected, and the voltage trend distribution is generated.

[0099] The current thermal electric synchronous determination sub-module compares the current fluctuation characteristics of the voltage trend distribution in the corresponding period with the thermal electric abnormal signal, judges the synchronism of the current and the thermal electric signal, selects the synchronous abnormal data segment, and generates the synchronous abnormal distribution;

[0100] The selected abnormal module numbers are called, the current monitoring sequence of each number in the voltage trend abnormal period is searched, the current data and the thermal electric abnormal signal are compared one by one, for each time point, whether the current fluctuation amplitude and the thermal electric abnormal response are synchronous is directly judged, by setting a threshold value, such as the current fluctuation absolute value exceeding 0.3A and the thermal electric abnormal signal appearing at the same time, then it is determined that the synchronization is abnormal, if the two are not synchronized, the data segment is eliminated. In the example, if the voltage trend of B module is abnormal from 8:00 to 8:05, the current data fluctuation amplitudes are 0.35A, 0.37A, 0.32A, 0.29A and 0.4A, and the thermal electric abnormal signal is also in the active state at the same time, then the 5-minute data segment is marked as synchronous abnormal. After the pairing determination of all abnormal module numbers and corresponding time period current and thermal electric signal is completed, all synchronous abnormal data segments are summarized, and the synchronous abnormal distribution is generated.

[0101] The array risk grading sub-module calculates the abnormal distribution of the components in the monitoring log and the climate data fluctuation interval involved in the synchronous abnormal distribution, analyzes the synchronous distribution range and abnormal data characteristics of each component, optimizes the data corresponding relationship, and obtains the array risk grading index.

[0102] According to the component number and abnormal period in the synchronous abnormal distribution, the running abnormal data of the corresponding period in the monitoring log is extracted, the meteorological database is further searched, the climate parameters such as environmental temperature, irradiance and wind speed in the same period are extracted, the overlapping degree of each component abnormal occurrence time and climate fluctuation interval is compared in turn, the interval in which the climate parameter fluctuation amplitude exceeds the set threshold and the abnormal data synchronously appears is marked separately, the distribution range of the synchronous abnormal data segment of each component in all monitoring periods is counted, the length of time and the number of abnormal segments covered by each component are counted, and then according to the correspondence between the number abnormal coverage rate and the climate fluctuation, the risk classification threshold is set, such as the component with abnormal coverage rate greater than 10% and climate fluctuation amplitude higher than 1.5 times of the historical average is classified as high risk, the abnormal coverage rate is between 5% and 10% and is classified as medium risk, and the abnormal coverage rate is less than 5% and is classified as low risk, in the example, the cumulative time of the synchronous abnormal segment of the C component in the monitoring period is 30 minutes, accounting for 12% of the total time, and the wind speed and temperature fluctuation in this segment are higher than the set standard, then the C component is judged as high risk, all component numbers, risk classification, distribution interval and climate influence parameters are sorted and archived, and the array risk classification index is obtained.

[0103] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application still belongs to the protection scope of the present application.

Claims

1. A cloud computing based photovoltaic performance evaluation system characterized in that, The system comprises: The electrical thermal response interpretation module is based on a cloud computing platform, analyzes the current drop amplitude and temperature rise speed data uploaded by the photovoltaic module, calculates the ratio and compares it with the same period of the previous day, identifies the trend of the ratio change, filters the key time period, performs time alignment, and obtains the synchronous thermal-electric abnormality characteristics; The abnormal aggregation distribution module is based on the synchronous thermal-electric abnormality characteristics, analyzes the abnormal segment distribution of the components in the region, calculates the cumulative number of abnormal segments, judges the aggregation degree, filters the key number interval, optimizes the correspondence between the number and the spatial coordinates, and generates the regional abnormal aggregation distribution characteristics; The abnormal aggregation distribution module comprises: The number-space matching sub-module is based on the synchronous thermal-electric abnormality characteristics, analyzes the correspondence between the photovoltaic module number and the spatial position, identifies the number of each photovoltaic module, performs data pairing of the number and the spatial coordinates, and obtains the number-space correspondence; The abnormal cumulative calculation sub-module is based on the number-space correspondence, calculates the total amount of abnormal segments of each number within the spatial range, analyzes the spatial distribution of the cumulative number of abnormal segments, compares the cumulative difference under each number, and judges the abnormal segment distribution trend of the same regional number to obtain the total number of abnormal segment accumulations; The aggregation interval screening sub-module screens the cumulative number in each number interval according to the total number of abnormal segment accumulations, compares the spatial distribution data of the number interval, calculates the continuous distribution status of the number interval abnormal segments, calculates the number interval abnormal aggregation degree, and generates the regional abnormal aggregation distribution characteristics; The rhythm deviation trajectory module is based on the regional abnormal aggregation distribution characteristics, analyzes the continuous multi-day current output of the photovoltaic module, compares the current sequence with the historical same period, calculates the current deviation trajectory, judges the continuous deviation section and its direction change, combines the component number marker position, and outputs the current deviation dynamic characteristics; The rhythm deviation trajectory module comprises: The current sequence acquisition sub-module is based on the regional abnormal aggregation distribution characteristics, analyzes the collected continuous multi-day per-minute current data of the photovoltaic module, combines the component number index, arranges the daily current data in time sequence, compares the data integrity and continuity through daily classification and archiving operations, and obtains the current sequence set; The deviation section identification sub-module calls the current sequence set, calculates the absolute difference between the current running day per-minute current and the historical same period per-minute current, screens the minute section of the daily continuous deviation, judges the continuous deviation and direction change, calculates the deviation section feature value, and obtains the current deviation section; The current deviation feature output sub-module analyzes the section range of the current deviation section, optimizes the component number index, compares the section feature with the actual position of the component, integrates each type of data, and outputs the current deviation dynamic characteristics; The voltage cumulative deviation module is based on the current deviation dynamic characteristics, analyzes the node voltage data, calculates the voltage change trend, screens the voltage segment with continuous deviation and cumulative change characteristics, judges whether the cumulative change is abnormal, and obtains the voltage cumulative deviation feature.

2. The cloud computing based photovoltaic performance evaluation system as claimed in claim 1, wherein, The synchronous thermal-electric anomaly feature includes an anomaly response ratio type, a correlation voltage change feature, and a key time label, the regional anomaly aggregation distribution feature includes an aggregation distribution type, a spatial density, and a numbered group identification result, the current offset dynamic feature includes a sustained offset category, a change direction attribute, and numbered mapping information, and the voltage cumulative offset feature includes a cumulative amplitude attribute, a classification result, and a node response grouping.

3. The cloud computing based photovoltaic performance evaluation system as claimed in claim 1, wherein, The electrical thermal response interpretation module includes: The ratio sequence calculation submodule is based on a cloud computing platform, analyzes current drop amplitude and temperature rise speed data uploaded by the photovoltaic module, calculates a ratio of the current drop amplitude and the temperature rise speed at each time point, compares ratio changes at consecutive time points, and obtains a ratio sequence trend; The key time period screening submodule judges change trends of the ratio sequence in each time period according to the ratio sequence trend and a ratio sequence of the same time period of the previous day, screens time periods in which the ratio changes significantly, and obtains a ratio change key interval; The thermal-electric anomaly feature extraction submodule analyzes voltage changes in the corresponding time period based on the ratio change key interval, judges synchronous features of the ratio change and the voltage change, performs time alignment, and obtains a synchronous thermal-electric anomaly feature.

4. The cloud computing based photovoltaic performance evaluation system as claimed in claim 1, wherein, The voltage cumulative offset module includes: The voltage trend calculation submodule analyzes voltage data continuously collected at each node based on the current offset dynamic feature, calculates a voltage change amplitude of each node at adjacent collection time points, judges continuous change conditions of the voltage at each collection stage, analyzes voltage fluctuation cumulative trends at each stage, and obtains a voltage change trend; The sustained offset screening submodule judges whether voltage increase and decrease directions at each collection stage in the voltage change trend are consistent, analyzes coverage intervals and direction continuity of sustained offset stages, compares offset direction features between difference segments, screens typical segments of the voltage sustained offset, and obtains sustained offset segments; The segment anomaly judgment submodule analyzes voltage change intervals of the sustained offset segments, calculates continuity and fluctuation distribution conditions of voltage change directions in each segment, judges whether the change performance deviates from a regular state, and obtains a voltage cumulative offset feature.

5. The cloud-computing based photovoltaic performance evaluation system according to claim 1, wherein, The system further includes: The risk early warning judgment module matches the current offset dynamic feature and the synchronous thermal-electric anomaly feature in the same time period based on the voltage cumulative offset feature, judges synchronous distribution of abnormal segments in time and module numbers, analyzes monitoring logs and combines climate fluctuation data, and obtains an array risk grading index; The array risk grading index includes a risk level category, a risk interval identifier, and an associated influencing factor.

6. The cloud-computing-based photovoltaic performance evaluation system according to claim 5, wherein, The risk early warning judgment module includes: The voltage offset detection submodule analyzes voltage change curves of the photovoltaic module in the same monitoring period based on the voltage cumulative offset feature, judges differences in voltage fluctuation trends between modules, screens module numbers with abnormal change trends, and generates a voltage trend distribution; The current thermal-electric synchronous determination submodule compares current fluctuation characteristics and thermal-electric anomaly signals of time periods corresponding to the voltage trend distribution, judges synchronicity of the current and the thermal-electric signals, screens data segments with synchronous anomalies, and generates a synchronous anomaly distribution; The array risk grading submodule calculates the synchronous abnormal distribution, which involves the abnormal distribution of components in the monitoring log and the climate data fluctuation interval, analyzes the synchronous distribution range of each component and the abnormal data characteristics, optimizes the data correspondence relationship, and obtains the array risk grading index.

Images