A method and system for photovoltaic array shadowing condition detection

Abstract

This invention relates to the field of pattern recognition technology, specifically to a method and system for detecting the shading state of a photovoltaic array. The method includes the following steps: determining output compression caused by the shift of the maximum power point; constructing a perturbation distribution through voltage sorting deviation; identifying branches of abrupt changes in power difference trends; extracting abnormal power decrease points to establish shading evolution monitoring; and combining temperature rise and power trends to determine response consistency and state verification. In this invention, by constructing a compression degree characterization, combining voltage output sequence perturbation analysis and power difference trend identification, a continuous evolution process expression mechanism for shading signals is established in the physical deployment dimension. Furthermore, by combining the directional matching relationship between temperature and power changes, multi-dimensional perception and cross-verification of the shading state are achieved, improving the local identification and dynamic monitoring capabilities of the shading-affected area, avoiding judgment interference caused by environmental fluctuations and non-shading anomalies, and enhancing stability and robustness under complex operating conditions.

Description

Technical Field

[0001] This invention relates to the field of pattern recognition technology, and in particular to a method and system for detecting the shading status of photovoltaic arrays. Background Technology

[0002] Pattern recognition technology involves the analysis, learning, and classification of various data patterns. This includes feature extraction, feature selection, similarity measurement, and classification decisions for inputs such as signals, images, speech, text, and various sensor data. The goal is to achieve automatic identification and discrimination of target objects or states. This encompasses applications based on statistical learning methods, neural network models, support vector machines, cluster analysis, principal component analysis, and other algorithms. It is applied in various scenarios such as biometrics, image recognition, fault diagnosis, medical testing, and industrial inspection. It is suitable for scenarios requiring the mapping of input data to several category labels, such as health status classification, object recognition, or environmental change detection. Traditional photovoltaic array shading state detection methods refer to... Methods for analyzing parameter changes during the operation of photovoltaic arrays to determine whether they are in a shading state typically target the phenomenon that the electrical output characteristics of photovoltaic modules change when they are shaded. This is done by collecting operating data such as voltage, current, power, or IV curves of photovoltaic cells and combining them with specific rules or empirical models to identify the shading state. Specifically, this includes determining the degree of shading based on the comparison of power change trends with the level of sunlight, identifying the shading of some modules by analyzing the output differences between photovoltaic modules in series, and determining the area of ​​light shading by collecting infrared or visible light images and using changes in image brightness and contrast. Traditional methods are mostly based on empirical threshold judgments or structured rule processing, and the identification process is greatly affected by environmental disturbances and has limited accuracy.

[0003] Traditional photovoltaic array shading status detection methods rely on direct comparisons between parameters such as power and voltage and empirical rules. In practical applications, these methods are affected by multiple factors such as changes in natural sunlight, component aging, and temperature fluctuations, causing the single threshold judgment mechanism to fail. They cannot effectively separate the essential differences between output anomalies caused by shading and other operational fluctuations. They lack the ability to model detailed response behaviors caused by shading structures, making it difficult to accurately depict the spatial distribution and evolution trend of shading interference. Electrical characteristic analysis remains at the average or interval level, ignoring the ordering relationship of output data in the array's physical layout. They cannot achieve fine-grained and dynamic identification and location of shading status changes, resulting in misjudgments and omissions. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and to propose a method and system for detecting the shading status of photovoltaic arrays.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for detecting the shading status of a photovoltaic array, comprising the following steps:

[0006] S1: Call the electrical measurement data of the photovoltaic array, analyze the open-circuit voltage and short-circuit current endpoints in the IV curve, calculate the available output range, determine the offset of the maximum power point on the voltage axis and current axis, and determine the degree of compression by comparing with the power point position under rated conditions, and obtain the compression offset strength coefficient.

[0007] S2: Based on the compression offset strength coefficient, compare the voltage output of each series branch, sort them, calculate the difference between the current sort and the reference sort, construct the offset sequence, and determine the clustering and change direction in the physical arrangement order to obtain the voltage order disturbance distribution.

[0008] S3: Using the voltage order disturbance distribution, calculate the power difference between adjacent branches and construct a power difference sequence. Analyze the changes in the difference in the sequence, identify the turning point of the power change trend, and number and mark the corresponding branches of the trend turning point to obtain the shielding level turning branch sequence number.

[0009] S4: Call the shielding level turning branch sequence number, locate the trend turning branch and the power decrease change trajectory of adjacent deployment areas, extract the abrupt change node and select the node with the largest power decrease as the shielding response benchmark point, monitor the stage changes, judge the shielding evolution trend, and establish a shielding status monitoring record.

[0010] S5: Call the shading state monitoring record, obtain the temperature sampling sequence and power output sequence corresponding to the component during the monitoring period, make a corresponding judgment on the direction and magnitude of the change in each point in the temperature rise change sequence and power decay change sequence, analyze the correspondence between the shading state and the temperature and electricity response, verify the shading state, and obtain the shading response judgment result.

[0011] As a further aspect of the present invention, the compression offset intensity coefficient includes voltage direction offset amplitude, current direction offset amplitude, and electrical working area coverage ratio; the voltage order disturbance distribution includes branch voltage sorting difference sequence, sequence offset cluster segment, and offset direction change characteristics; the shielding level turning branch sequence number specifically includes the power difference change abrupt change position number, power change trend inflection point number, and branch physical layout corresponding number; the shielding status monitoring record includes power decrease segment change trajectory, abrupt change node amplitude sequence, and trend continuity time marker; and the shielding response judgment result specifically refers to the thermal response trend sequence, electrical response trend sequence, and thermoelectric response correspondence analysis result.

[0012] As a further aspect of the present invention, the step of obtaining the compression offset strength coefficient specifically includes:

[0013] S111: Acquire electrical measurement data of photovoltaic array, collect open-circuit voltage endpoint and short-circuit current endpoint of IV curve, calculate the interval length in voltage direction and current direction, determine the length of available output interval based on endpoint difference, and generate available output interval value;

[0014] S112: Based on the available output range value, determine the offset position of the maximum power point on the voltage axis and the current axis, call the maximum power point coordinates calibrated under the rated operating state, and compare the differences in the corresponding directions of the voltage axis and the current axis coordinates to obtain the maximum power point offset.

[0015] S113: Based on the maximum power point offset, analyze the impact of the offset behavior on the coverage range of the available output interval, determine the degree of compression of the output interval coverage range by the offset behavior, construct the masking structure interference characteristics, and obtain the compression offset intensity coefficient.

[0016] As a further aspect of the present invention, the step of obtaining the voltage order disturbance distribution is specifically as follows:

[0017] S211: Based on the compression offset strength coefficient, compare the voltage output values ​​of each branch of the photovoltaic array at the same time, sort the branches according to the output magnitude, and generate a branch voltage sorting sequence.

[0018] S212: Call the branch voltage sorting sequence, calculate the position difference between the current sorting position and the preset reference position, record the sorting offset of each branch, and establish a voltage sorting offset sequence;

[0019] S213: Based on the voltage sorting offset sequence, by performing continuity analysis, determine the aggregation characteristics and change direction of sorting differences in the physical arrangement order of branches, identify local disturbance sections of the sorting structure, and generate voltage order disturbance distribution.

[0020] As a further aspect of the present invention, the step of obtaining the shielding level transition branch sequence number specifically includes:

[0021] S311: Based on the voltage order disturbance distribution, call the power output values ​​of adjacent branches of the photovoltaic array, perform difference calculation on physically adjacent branches, record the power difference between each adjacent branch according to the deployment order, and generate a branch power difference sequence.

[0022] S312: Call the branch power difference sequence, compare and calculate the adjacent differences in the sequence, analyze the increase change between adjacent differences, determine the abnormal position of the difference amplitude change in the power difference sequence, record the turning point of the power difference change, and obtain the turning point value of the power difference sequence.

[0023] S313: Based on the power difference sequence turning point value, and based on the output change direction and amplitude characteristics of the branches on both sides of the power trend turning point, the corresponding branches of the trend turning point are numbered and marked to generate the shading level turning point branch sequence number.

[0024] As a further aspect of the present invention, the step of obtaining the shading state monitoring record specifically includes:

[0025] S411: Based on the shielding level turning branch sequence number, call the power output continuous monitoring sequence, obtain the power output data of the trend turning branch and adjacent deployment areas, analyze the change trajectory of power output in the decreasing section, identify and locate the sudden node during the decreasing process, and obtain the sudden node location result.

[0026] S412: Based on the mutation node location result, call the power output data corresponding to the mutation node, compare the power difference of each node in the falling segment, select the node with the largest power difference as the shading response reference point, calculate the power change amplitude and time difference of each sampling point relative to the reference point in the continuous monitoring sequence, construct the change rate sequence corresponding to the reference point, calculate the change rate influence value associated with the reference point, and obtain the reference power change trend value.

[0027] S413: Call the reference power change trend value, analyze the change direction of the power difference sequence of the sampling points in the monitoring section, determine the stage structure between the power change direction and change amplitude in the continuous time segment, form the evolution trajectory of the shading state in the time dimension, and obtain the shading state monitoring record.

[0028] As a further aspect of the present invention, the step of obtaining the occlusion response judgment result specifically includes:

[0029] S511: Call the shading state monitoring record, obtain the temperature sampling sequence corresponding to the component during the monitoring period, perform difference judgment based on the temperature difference between adjacent sampling points in the sequence, construct the temperature rise change trajectory using the difference direction and difference amplitude, generate a change magnitude sequence based on the continuity of the trajectory in the time sequence, and obtain the temperature rise change amplitude.

[0030] S512: Based on the temperature rise change magnitude, call the power output sequence of the component during the monitoring period, construct the power attenuation change trajectory based on the power difference between adjacent sampling points, analyze the change magnitude of the trajectory in the time sequence, make a correspondence judgment on the change direction and magnitude of each point in the temperature rise change sequence and the power attenuation change sequence, and calculate the response matching relationship value.

[0031] S513: Based on the response matching relationship value, determine the correspondence of the shading state at the thermoelectric response level, verify the shading state, and obtain the shading response judgment result.

[0032] A system for detecting the shading status of a photovoltaic array, the system being used to perform the aforementioned method for detecting the shading status of a photovoltaic array, the system comprising:

[0033] The parameter range analysis module calls the electrical measurement data of the photovoltaic array, analyzes the open-circuit voltage and short-circuit current endpoints in the IV curve, calculates the available output range, determines the offset of the maximum power point on the voltage axis and current axis, and determines the degree of compression by comparing it with the power point position under rated conditions, and obtains the compression offset strength coefficient.

[0034] The branch order sequencing module compares the voltage output of each branch according to the compression offset intensity coefficient, calculates the difference between the current sort and the reference sort after sorting, constructs the offset sequence, and determines the aggregation and change direction in the physical arrangement order to obtain the voltage order perturbation distribution.

[0035] The power transition identification module uses the voltage order disturbance distribution to calculate the power difference between adjacent branches and construct a power difference sequence. It analyzes the changes in the difference in the sequence, identifies the turning point of the power change trend, and numbers and marks the corresponding branches of the trend turning point to obtain the shielding level turning branch sequence number.

[0036] The shading evolution tracking module calls the shading level turning branch sequence number, locates the trend turning branch and the power decrease change trajectory of adjacent deployment areas, extracts abrupt change nodes and selects the node with the largest power decrease as the shading response benchmark point, monitors the stage changes, judges the shading evolution trend, and establishes a shading status monitoring record.

[0037] The temperature and electrical consistency determination module calls the shading state monitoring record to obtain the temperature sampling sequence and power output sequence corresponding to the component during the monitoring period. It makes a corresponding judgment on the direction and magnitude of the change in each point in the temperature rise change sequence and power decay change sequence, analyzes the correspondence between the shading state and the temperature and electrical response, verifies the shading state, and obtains the shading response judgment result.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0039] In this invention, by constructing a compression degree characterization, combining voltage output sequence perturbation analysis and power difference trend identification, a continuous evolution process expression mechanism for the shielding signal is established in the physical deployment dimension. Combined with the directional matching relationship between temperature and power changes, multi-dimensional perception and cross-verification of the shielding state are achieved, improving the local identification and dynamic monitoring capabilities of the shielding-affected area, avoiding judgment interference caused by environmental fluctuations and non-shielding anomalies, and enhancing stability and robustness under complex operating conditions. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the workflow of the present invention;

[0041] Figure 2 This is a flowchart of the process for obtaining the compression offset strength coefficient of the present invention;

[0042] Figure 3 This is a flowchart of the voltage order disturbance distribution acquisition process of the present invention;

[0043] Figure 4 This is a flowchart of the process for obtaining the shielding level transition branch sequence number in this invention.

[0044] Figure 5 This is a flowchart of the process for acquiring the shading status monitoring record according to the present invention;

[0045] Figure 6 This is a flowchart of the process for obtaining the occlusion response judgment result of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0047] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] Please see Figure 1 This invention provides a technical solution, a method for detecting the shading status of a photovoltaic array, comprising the following steps:

[0049] S1: Call the electrical measurement data of the photovoltaic array, analyze the open-circuit voltage and short-circuit current endpoints in the IV curve, calculate the available output range, determine the offset of the maximum power point on the voltage axis and current axis, and determine the degree of compression by comparing with the power point position under rated conditions, and obtain the compression offset strength coefficient.

[0050] S2: Based on the compression offset strength coefficient, compare the voltage output of each series branch, sort them, calculate the difference between the current sort and the reference sort, construct the offset sequence, and determine the clustering and change direction in the physical arrangement order to obtain the voltage order disturbance distribution.

[0051] S3: Using the voltage order disturbance distribution, calculate the power difference between adjacent branches and construct a power difference sequence. Analyze the changes in the difference in the sequence, identify the turning point of the power change trend, and number and mark the corresponding branches at the trend turning point to obtain the branch sequence number of the shading level turning point.

[0052] S4: Call the shading level turning point branch sequence number, locate the trend turning point branch and the power decrease change trajectory of adjacent deployment areas, extract the abrupt change node and select the node with the largest power decrease as the shading response benchmark point, monitor the phased changes, judge the shading evolution trend, and establish a shading status monitoring record.

[0053] S5: Call the shading status monitoring record, obtain the temperature sampling sequence and power output sequence corresponding to the component during the monitoring period, make a corresponding judgment on the direction and magnitude of the change in temperature rise sequence and power decay sequence, analyze the correspondence between the shading status and the temperature and electricity response, verify the shading status, and obtain the shading response judgment result.

[0054] The compression offset intensity coefficient includes the voltage direction offset amplitude, the current direction offset amplitude, and the coverage ratio of the electrical working area. The voltage order disturbance distribution includes the branch voltage sorting difference sequence, the sequence offset cluster segment, and the offset direction change characteristics. The shielding level turning point branch sequence number specifically includes the power difference change abrupt change position number, the power change trend inflection point number, and the corresponding number of the branch physical layout. The shielding status monitoring record includes the power drop segment change trajectory, the abrupt change node amplitude sequence, and the trend continuity time mark. The shielding response judgment result specifically refers to the thermal response trend sequence, the electrical response trend sequence, and the thermoelectric response correspondence analysis result.

[0055] Please see Figure 2 The specific steps for obtaining the compressive offset strength coefficient are as follows:

[0056] S111: Acquire electrical measurement data of photovoltaic array, collect open-circuit voltage endpoint and short-circuit current endpoint of IV curve, calculate the interval length in voltage direction and current direction, determine the length of available output interval based on endpoint difference, and generate available output interval value;

[0057] To acquire electrical measurement data of the photovoltaic array, high-precision DC voltage transformers and Hall current sensors deployed on the photovoltaic combiner box side are used to synchronously collect real-time IV characteristic scan data of the entire photovoltaic array at a preset sampling time (e.g., 10:00:00 AM). The open-circuit voltage endpoint is extracted from the IV curve dataset obtained from the scan. With short-circuit current endpoint Set the voltage measurement zero-point reference as The zero-point reference for current measurement is set as Calculate the physical length of the interval in the voltage direction, i.e., perform the operation. Input the collected instance measurement values Calculations yielded Calculate the physical length of the interval in the direction of the current, i.e., perform the operation. Input the collected instance measurement values Calculations yielded Based on the voltage terminal difference Difference between current terminals These two physical quantities are jointly determined as the effective working boundary of the current photovoltaic array in the electrical dimension, generating a usable output range value. ).

[0058] S112: Based on the available output range values, determine the offset position of the maximum power point on the voltage axis and current axis, call the maximum power point coordinates calibrated under the rated operating state, compare the differences in the corresponding directions of the voltage axis and current axis coordinates, and obtain the maximum power point offset.

[0059] Based on the available output range values ​​( ), within a defined voltage domain With the current domain The maximum power point tracking (MPPT) algorithm is performed internally to identify the maximum power point under the current actual operating conditions. and its corresponding voltage coordinates With current coordinates To obtain a specific instance measurement value, where and It retrieves the rated operating status parameters stored in the system database, which are based on the standard test condition STC (irradiance). Battery temperature The maximum power point coordinates calibrated under AM1.5 atmospheric quality are as follows: and The voltage axis coordinates are compared according to the differences in the direction of execution, and the absolute voltage deviation is calculated. The difference in the direction of the current axis coordinate is compared to calculate the absolute deviation of the current. Obtain the maximum power point offset ( ).

[0060] S113: Based on the maximum power point offset, analyze the impact of the offset behavior on the coverage of the available output range, determine the degree of compression of the output range coverage by the offset behavior, construct the interference characteristics of the shielding structure, and obtain the compression offset intensity coefficient.

[0061] Based on the maximum power point offset and First, we analyze the impact of voltage offset behavior on the compression of the available output range coverage, and then call the rated voltage coordinates obtained in step S112. Calculate voltage offset rate Next, the impact of current offset behavior on the compression of the available output range is analyzed, and the rated current coordinates obtained in step S112 are called up. Calculate the current offset rate Determine the degree to which the offset behavior compresses the output range coverage, and adjust the voltage offset rate. With current offset All are considered as characteristic components representing the distortion of the IV curve caused by shading. A shading structure interference characteristic model is constructed, and the comprehensive intensity of this characteristic is calculated by weighted summation, with voltage offset weights set. and current offset weight All The weight value is set based on the photovoltaic power output formula. The equal importance of voltage and current in the power contribution of the calculation The compressive offset strength coefficient is obtained. The numerical result indicates that the output characteristics of the current photovoltaic array have changed by approximately [a certain percentage] compared to the ideal rated state. The comprehensive compression offset indicates a potential anomalous state.

[0062] Please see Figure 3 The specific steps for obtaining the voltage order disturbance distribution are as follows:

[0063] S211: Based on the compression offset strength coefficient, compare the voltage output values ​​of each branch of the photovoltaic array at the same time, sort the branches according to the output magnitude, and generate a branch voltage sorting sequence.

[0064] Based on the compressive offset strength coefficient First, set a threshold for determining whether to trigger an occlusion. The threshold The setup was achieved by acquiring historical data from the array under standard test conditions (STC) and under multiple uniform but non-rated low-light conditions (such as early morning, evening, and uniform thin clouds). The data was statistically analyzed, and the mean of these 100 sets of unshaded state data was calculated. with standard deviation ,Pick As the upper limit of the confidence interval, this upper limit is calculated to be To enhance anti-interference capabilities and reserve redundancy, the final setting was... Execute the decision logic In the example If the result is true, confirming the risk of abnormal shading, a comparison program is initiated to evaluate the voltage output values ​​of each branch of the photovoltaic array at the same time. This is achieved by simultaneously reading the MPP terminal voltage values ​​of 10 branches (numbered SL1 to SL10) at the same time (e.g., 10:00:05 AM) using a multi-channel data acquisition card. (in The obtained voltage data set (units) The output values ​​are: [552(SL1),550(SL2),551(SL3),549(SL4),480(SL5),475(SL6),482(SL7),550(SL8),548(SL9),553(SL10)]. Based on the rule of output values ​​from largest to smallest, these 10 branches are sorted in descending order to generate the branch voltage sorting sequence [SL10(553V),SL1(552V),SL3(551V),SL2(550V),SL8(550V),SL4(549V),SL9(548V),SL7(482V),SL5(480V),SL6(475V)].

[0065] S212: Call the branch voltage sorting sequence, calculate the position difference between the current sorting position and the preset reference position, record the sorting offset of each branch, and establish a voltage sorting offset sequence;

[0066] Call the branch voltage sorting sequence (denoted as the current position) The actual ranking of each branch in this sequence is as follows: SL10 1st, SL1 2nd, SL3 3rd, SL2 4th, SL8 5th, SL4 6th, SL9 7th, SL7 8th, SL5 9th, SL6 10th. Simultaneously, the preset reference order stored in the system configuration table is invoked. This reference sequence is based on the electrical topology, cable impedance differences, and component factory parameter consistency determined during the photovoltaic array design and installation phase. It is also the ideal sequence determined by calculating the statistical average of voltage data collected at the same time (10:00 AM) on seven consecutive sunny, unshaded days during the system grid-connected commissioning phase. This reference sequence is set as [SL1 (1st position), SL2 (2nd position), SL3 (3rd position), SL4 (4th position), SL5 (5th position), SL6 (6th position), SL7 (7th position), SL8 (8th position), SL9 (9th position), SL10 (10th position)]. The current sequence position is calculated... With preset reference sequence The positional differences between them, that is, for each branch road Perform operation The specific calculation is as follows: SL1 offset SL2 offset SL3 offset SL4 offset SL5 offset SL6 offset SL7 offset SL8 offset SL9 offset SL10 offset Record all the above calculation results and establish a voltage sorting offset sequence. .

[0067] S213: Based on the voltage sorting offset sequence, through continuity analysis, determine the aggregation characteristics and change direction of sorting differences in the physical arrangement order of branches, identify local disturbance sections of the sorting structure, and generate voltage order disturbance distribution.

[0068] Offset sequence sorted by voltage By this sequence A continuity analysis is performed based on the physical arrangement order of the branches (i.e., from physical position SL1 to physical position SL10) to determine the clustering characteristics of the sorting differences in the physical arrangement order of the branches, and a threshold for offset clustering is set. The threshold is set based on historical unmasked data. The statistical characteristics of the series fluctuation amplitude are selected, with the upper limit of the 95% confidence interval chosen (e.g., (bit) is used as the boundary for normal fluctuations, therefore, it is set Scan the sequence one by one Identify absolute values ​​greater than The elements were discovered. and The offsets of two physically consecutive branches are both significantly greater than At the same time, analyze its direction of change. and All are positive offsets (a positive value indicates a shift in rank in descending order, i.e., a relative decrease in voltage), and their physical neighbors... Although it did not exceed the threshold, it showed a positive offset trend in the same direction. It was found that there was a local perturbation in the sorting structure in the physically arranged SL5-SL6-SL7 section. This perturbation was manifested as a collective and significant decrease in voltage ranking relative to the normal reference state, generating a voltage order perturbation distribution. This distribution marked the physically continuous branch section from SL5 to SL7 as a key suspected shielding area of ​​voltage collapse.

[0069] Please see Figure 4 The specific steps for obtaining the shading level transition branch sequence number are as follows:

[0070] S311: Based on the voltage order disturbance distribution, call the power output values ​​of adjacent branches of the photovoltaic array, perform difference calculation on physically adjacent branches, record the power difference between each adjacent branch according to the deployment order, and generate a branch power difference sequence.

[0071] Based on the voltage order disturbance distribution (marked key areas SL5-SL7), the system retrieves the real-time power output values ​​of all physically adjacent branches in the photovoltaic array. (in This was used to accurately verify the power of voltage anomaly regions and their boundaries, obtaining instance power data (unit: ...). As shown in Table 1 below:

[0072] Table 1 Branch Power Output Data Table

[0073]

[0074] As shown in Table 1, the system considers physically adjacent branches (i.e., and Perform difference calculations, the formula is as follows: Calculate the power difference between each pair of adjacent branches in the order of physical deployment (SL1 and SL2, SL2 and SL3, ..., SL9 and SL10). The specific calculation process is as follows: , , , , , , , , Arrange these differences in order to generate a branch power difference sequence. .

[0075] S312: Call the branch power difference sequence, compare and calculate the adjacent differences in the sequence, analyze the increase change between adjacent differences, determine the abnormal position of the difference amplitude change in the power difference sequence, record the turning point of the power difference change, and obtain the turning point value of the power difference sequence.

[0076] Calling the branch power difference sequence For the sequence Difference between adjacent physical quantities and Perform second-order comparison operations to analyze the changes in the increase between adjacent differences, specifically by executing the formula. The sequence of increases between adjacent differences was calculated. (unit The calculation process is as follows: , , , , , , , Set an abnormal threshold for the magnitude of the difference. This threshold is based on the array under uniform illumination. The series' historical maximum statistical volatility (e.g., 100W) is used as the basis for joint calibration, along with a safety margin factor (e.g., 400W), to ultimately set... Execute the judgment logic to filter the power difference sequence. The location, in In the sequence, identify , , , All greater than Record the power difference change inflection points corresponding to these abnormal increases, obtain the inflection point values ​​of the power difference sequence, and accurately pinpoint the location. (correspond (the resulting mutations) and (correspond (The resulting mutations).

[0077] S313: Based on the power difference sequence turning point value, and based on the output change direction and amplitude characteristics of the branches on both sides of the power trend turning point, the corresponding branches of the trend turning point are numbered and marked to generate the shielding level turning point branch sequence number.

[0078] Based on the power difference sequence inflection point value ( and ), and perform detailed output feature analysis on the branches on both sides of the identified turning point. First, analyze The corresponding transition segment, the power difference sequence of this segment from violent mutation ,according to The definition identifies the power output of branch SL5. Compared to its preceding physically adjacent branch SL4 A significant drop occurred (amplitude characteristics meet anomaly criteria), and the direction of change was negative (directional characteristic). Based on this, SL5 was determined to be the first branch entering the shielded area. Further analysis followed. The corresponding transition segment, the power difference sequence of this segment from violent mutation ,according to The definition identifies the power output of branch SL8. Compared to its preceding physically adjacent branch SL7 A significant increase occurred (the amplitude characteristic meets the abnormal criteria), and the direction of change was positive (directional characteristic). Based on this, SL7 was determined to be the last branch in the shading zone. The corresponding branches at these two power trend inflection points (between SL4 and SL5, and between SL7 and SL8) were numbered and marked. The branch SL5, which was the starting branch where the significant decrease occurred, was marked as the "shading start inflection point," and the branch SL7, which was the preceding branch where the significant increase occurred, was marked as the "shading end inflection point." The branch sequence number of the shading level inflection point was generated, and the final marking result was the set {SL5,SL7}.

[0079] Please see Figure 5 The specific steps for obtaining the shading status monitoring record are as follows:

[0080] S411: Based on the shielding level turning branch sequence number, call the power output continuous monitoring sequence, obtain the power output data of the trend turning branch and adjacent deployment areas, analyze the change trajectory of power output in the decreasing section, identify and locate the sudden node during the decreasing process, and obtain the sudden node location result.

[0081] Based on the shielding level transition branch numbers {SL5,SL7}, the system's continuous power output monitoring sequence function is invoked to selectively collect real-time power output data for the trend transition branches SL5, SL7, and their adjacent deployment areas (SL4,SL6,SL8). Set the data sampling frequency to (That is, one data point is collected every 10 seconds). Taking branch SL5 as an example, its continuous power output data (unit: ) is obtained during the period when the shading event occurs (10:00:00 to 10:00:50). The sequence is: [4980(t=0s), 4950(t=10s), 4500(t=20s), 3500(t=30s), 3150(t=40s), 3100(t=50s)]. Analyze the dynamic trajectory of this power output in the decreasing region (t=10s to t=50s), and calculate the power difference between adjacent time sampling points. The power change sequence over time was obtained (units). [-30, -450, -1000, -350, -50], sets a mutation detection threshold. This threshold is based on the fastest rate of power decline caused by normal light fluctuations (such as the rapid passage of thin, soft clouds) in historical meteorological data (e.g., Calibration was performed, and the result was calculated based on a 10-second sampling interval. Traverse the sequence to identify and locate the conditions that satisfy the descent process. The mutation node under the condition was identified, and the mutation node location result was obtained, which was locked at time point t=30s (corresponding to...). , ).

[0082] S412: Based on the location results of the mutation nodes, retrieve the power output data corresponding to the mutation nodes, compare the power difference of each node in the descent segment, select the node with the largest power difference as the shading response reference point, calculate the power change amplitude and time difference of each sampling point relative to the reference point in the continuous monitoring sequence, construct the change rate sequence corresponding to the reference point, and use the formula:

[0083] ;

[0084] Calculate the influence value of the rate of change of the reference point to obtain the trend value of the reference power change;

[0085] in, This is the rate of change impact value associated with the reference point, used to measure the weighted trend strength between power change and time change at the reference point within a continuously monitored section. For the first The normalized power value at each sampling point is obtained by dividing the original power value by the rated power value. The power normalization value at the baseline point is obtained by selecting the node with the largest power decrease among the abrupt change nodes and normalizing it. For the first The normalized time values ​​of each sampling point are obtained by converting the sampling time values ​​into a zero-mean unit time series. The normalized time value for the reference point is obtained by normalizing the sampling time of the reference point. This is the index number of the current sampling point in the continuous monitoring sequence. This represents the total number of sampling points in the current power monitoring section;

[0086] Based on the mutation node location result (t=30s), retrieve the power output data corresponding to that mutation node. Given that in this monitoring section [4980,4950,4500,3500,3150,3100], only The absolute value of the decrease at this moment Exceeded the threshold Therefore, the system selects Corresponding nodes ( ) as the occlusion response reference point Then, the normalized parameter sequence required for the calculation is constructed. First, the rated power value of the branch is set. Perform a normalization operation $i=P_i / P{rated}$ on the power sequence to calculate the result. This determines the normalized power value at the reference point. Secondly, perform max-min normalization operation on the time series [0s, 10s, 20s, 30s, 40s, 50s]. (in ), calculated This determines the normalized time value of the reference point. Confirm the total number of sampling points in the current monitoring section. Using formula Calculate the influence value of the rate of change associated with the benchmark point. The calculation logic of this formula is explained in detail below: For each sampling point within the monitoring section... ( From 1 to The sampling point index, (Total number of sampling points), first calculate its distance from the reference point. The power-time change rate between (i.e.) ),in For the first The power normalization value of each sampling point The power normalization value at the reference point, For the first Normalized time values ​​for each sampling point The normalized time value is taken as the reference point, and the absolute value of the rate of change is used to measure the drasticness of the change. This is then multiplied by a weighting coefficient based on the power amplitude (i.e., The denominator of the weighting coefficient Follow The increase in power means that high-power points (in the early stages of occlusion) have relatively lower weights, while low-power points (in the deeper stages of occlusion) have relatively higher weights. The symbol represents all sampling points ( arrive The calculated weighted rate values ​​are accumulated. It should be noted that when... Make Time (i.e.) This term has a denominator of zero; in the calculation logic, this term is directly defined as having a value of zero. The advantage of the formula lies in the introduction of a reference point power. and current point power As a weighting factor, the calculation result It not only measures the rate of change of power over time, but also comprehensively considers the relative magnitude of the power value itself during the shading process, thus enabling it to more sensitively capture the dynamic evolution characteristics centered on the shading core (reference point) and perform detailed calculations using example data. ):

[0087] , : ;

[0088] , : ;

[0089] , : ;

[0090] , as the baseline, the value of this item is set to 0;

[0091] , : ;

[0092] , : ;

[0093] Perform accumulation operation ;

[0094] Calculate the influence of the rate of change of the reference point to obtain the trend value of the reference power change. The numerical result It is a dimensionless comprehensive index. The larger the value, the more drastic and concentrated the power change around the reference point. This value will be used in subsequent steps to quantitatively determine the type of shading (such as fast-moving shading or slow-moving shading).

[0095] S413: Call the baseline power change trend value, analyze the change direction of the power difference sequence of sampling points in the monitoring section, determine the stage structure between the power change direction and change amplitude in the continuous time segment, form the evolution trajectory of the shading state in the time dimension, and obtain the shading state monitoring record.

[0096] Call the reference power change trend value Simultaneously, the power difference sequence calculated in step S411 is invoked. (unit The difference sequence [-30, -450, -1000, -350, -50] was analyzed in detail for its direction of change (all negative values, indicating a continuous decline) and amplitude characteristics. The sequence was identified as exhibiting a typical three-segment structure: the first segment is a "gradual decline zone" (-30, -450), the second segment is a "rapid decline zone" (-1000), and the third segment is a "decelerated decline and stable zone" (-350, -50). Based on this, the power change phases within this continuous time period (t=0s to t=50s) were determined, forming the evolution trajectory of the shading state in the time dimension. Specifically, this trajectory is described as follows: the shading effect begins to appear at t=20s (abrupt change in power slope), the shading response reaches a sharp peak at t=30s (reference point), and after t=40s, the shading tends to reach a complete coverage state (entering a stable low-power region). A threshold for the rate of change intensity was set. The threshold is set based on different types of occlusion events in the historical database. Statistical analysis and calibration of the values ​​show that soft cloud cover... Values ​​are typically distributed in The area, while hard shadows cast by buildings or bird droppings obscure its view. Values ​​are typically distributed in The interval is selected accordingly. As a criterion for distinguishing between rapid and gradual occlusion, the calculation results will be used. With threshold Compare and determine Combining this evolutionary trajectory with The judgment result (classified as rapid mutation type) generates a shading status monitoring record. This record clearly marks that the SL5 branch experienced a "rapid mutation type" shading event between 10:00:00 and 10:00:50, and its peak influence point is located at t=30s.

[0097] Please see Figure 6 The specific steps for obtaining the occlusion response judgment result are as follows:

[0098] S511: Call the shading status monitoring record, obtain the temperature sampling sequence corresponding to the component during the monitoring period, perform difference judgment based on the temperature difference between adjacent sampling points in the sequence, construct the temperature rise change trajectory using the difference direction and difference amplitude, generate a change magnitude sequence based on the continuity of the trajectory in the time sequence, and obtain the temperature rise change amplitude.

[0099] The shading status monitoring record (for the SL5 branch, time window t=0s to t=50s) is invoked. Real-time temperature sampling sequences of key components on the SL5 branch during the monitoring period are obtained through thermocouples or infrared temperature sensors attached to the component backplane. (unit The collected temperature sequence The values ​​are: [45.0 (t=0s), 45.1 (t=10s), 45.5 (t=20s), 48.0 (t=30s), 50.5 (t=40s), 51.5 (t=50s)]. Based on adjacent sampling points in the sequence... and The temperature difference is used to perform a difference judgment, that is, to calculate... Using the direction of difference ( All values ​​are positive, indicating temperature rise. A temperature rise trajectory is constructed based on the difference in magnitude. A sequence of magnitude of change is generated based on the continuity of the trajectory in time sequence (t=10s to t=50s). (unit The range is: [0.1, 0.4, 2.5, 2.5, 1.0], which gives the magnitude of the temperature rise.

[0100] S512: Based on the temperature rise change magnitude, retrieve the component's power output sequence during the monitoring period. Construct a power attenuation trajectory based on the power difference between adjacent sampling points. Analyze the trajectory's change magnitude over time. Determine the correspondence between the direction and magnitude of change at each point in the temperature rise change sequence and the power attenuation change sequence using the following formula:

[0101] ;

[0102] Calculate the response matching value;

[0103] in, The response matching value is used to represent the consistency relationship between the temperature rise trajectory and the power decay trajectory. The first in the temperature rise trajectory The normalized temperature rise variation of the term is obtained by calculating the difference between adjacent points in the temperature sampling sequence and then performing max-min normalization. The first in the power decay change trajectory The normalized power attenuation magnitude of the term is obtained by calculating the difference between adjacent points in the power output sequence and then performing max-min normalization. The first in the temperature rise trajectory The trend offset of the term is extracted by determining the sign of the difference direction between adjacent points in the temperature rise trajectory and constructing a trend direction change sequence. The first in the power decay change trajectory The trend offset of the term is extracted by determining the sign of the difference direction between adjacent points in the power decay change trajectory and constructing a trend direction change sequence. The total number of valid sampling points involved in the corresponding judgments for the temperature rise trajectory and the power decay trajectory is obtained by taking the intersection of the number of temperature and power samples acquired synchronously during the monitoring period. This is a sampling point index variable used to mark the first sampling point in the sample sequence. One matching point;

[0104] Based on the magnitude of temperature rise The power output sequence of the SL5 branch during the monitoring period was invoked. (From S411), based on the power difference between adjacent sampling points (unit Construct a power decay trajectory using the formula [-30, -450, -1000, -350, -50], analyze the magnitude of the trajectory's change over time, and perform absolute value calculations. (unit Calculate the attenuation amplitude to obtain For the temperature rise change sequence With power attenuation amplitude sequence The direction of change of each point in the middle ( All are positive. All values ​​are negative, and their directions are opposite, which is consistent with the physical characteristics of the photovoltaic hot spot effect. To determine the correspondence between the normalized values ​​and the amplitude, first, the total number of valid sampling points is determined. Both sequences have 5 valid data points in the interval from t=10s to t=50s, therefore Secondly, as shown in Table 2, the two sequences are subjected to min-max normalization:

[0105] Table 2. Normalized data of temperature rise and power decay sequences.

[0106]

[0107] Then, the trend offset of the two normalized sequences is calculated. and ,definition and for and The deviation from the arithmetic mean of their respective sequences is first calculated. mean of the sequence ,calculate mean of the sequence Using formula Calculate the response matching value The calculation logic of this formula is explained in detail below: The total number of valid sampling points. From 1 to The sampling point index, and The first Normalized temperature rise and power decay at the point of view and These are their respective trend offsets; the core of the formula lies in calculation. That is, temperature rise and power decay in the first... The difference in normalized amplitude of points, denominator It is a normalized weighting factor that combines the volatility characteristics of both, used to measure the first... The degree to which point data deviates from the overall average trend. The symbol represents all The weighted difference ratios calculated from each sampling point are summed up. This indicates the calculation of this The advantage of the formula is that it calculates the average matching difference of each point. and The difference between them, and their deviation from the trend mean. Weighting is performed so that The value can sensitively reflect the similarity of the two curves in waveform shape (i.e., the synchronicity of the sharp drop in power leading to a sharp rise in temperature in the "hot spot effect"), rather than just the closeness of the numerical values. Detailed calculations can be performed by substituting the values ​​into Table 2 and the mean data. ):

[0108] : , molecule is The result is ;

[0109] : , This item ;

[0110] : , molecule is The result is ;

[0111] : , This item ;

[0112] : , This item ;

[0113] Perform summation and average operations Calculate the response matching relationship value to obtain The smaller this value, the better the match between the temperature rise curve and the power decay curve.

[0114] S513: Based on the response matching relationship value, determine the correspondence between the shading state and the thermoelectric response level, verify the shading state, and obtain the shading response judgment result.

[0115] Based on response matching relationship value Identify the degree of response matching of this value within the time period of change, and set a temperature-electricity matching judgment threshold. This threshold is based on "real hot spot shading events" (which are manually verified and calibrated from a large amount of historical field data) The mean is usually distributed in Between) and "non-physical shielding faults" (such as current sensor drift causing a drop in power readings but no significant change in component temperature, such cases) Values ​​are usually )of The value distribution characteristics are calibrated, and the following selections are made. As a benchmark for determining whether strong physical coupling (strong matching) exists, the calculated value will be... and Compare and perform judgment In the example The judgment logic is valid. This result shows that the temperature rise change and power attenuation have a very high consistency in the normalized trend. The temperature-electric matching coefficient is obtained and it is judged as "strong matching". This confirms that there is a clear physical correspondence between the shielding state and the temperature-electric response level. That is, it is confirmed that the power drop and temperature rise of the SL5 branch have a high degree of synchronicity and causal coupling. Based on this, the "rapid mutation type" shielding event in the record is finally verified and the shielding response judgment result is obtained. Finally, the system output confirms that "the SL5 branch has a real physical shielding".

[0116] A system for detecting the shading status of a photovoltaic array, the system being used to execute the aforementioned method for detecting the shading status of a photovoltaic array, the system comprising:

[0117] The parameter range analysis module calls the electrical measurement data of the photovoltaic array, analyzes the open-circuit voltage and short-circuit current endpoints in the IV curve, calculates the available output range, determines the offset of the maximum power point on the voltage axis and current axis, and determines the degree of compression by comparing it with the power point position under rated conditions, and obtains the compression offset strength coefficient.

[0118] The branch order sequencing module compares the voltage output of each branch according to the compression offset intensity coefficient, calculates the difference between the current sort and the reference sort after sorting, constructs the offset sequence, and determines the aggregation and change direction in the physical arrangement order to obtain the voltage order perturbation distribution.

[0119] The power transition identification module uses the voltage order disturbance distribution to calculate the power difference between adjacent branches and construct a power difference sequence. It analyzes the changes in the difference in the sequence, identifies the turning point of the power change trend, and numbers and marks the corresponding branches at the trend turning point to obtain the shielding level turning branch sequence number.

[0120] The shading evolution tracking module calls the shading level inflection branch sequence number, locates the trend inflection branch and the power decrease change trajectory of adjacent deployment areas, extracts abrupt change nodes and selects the node with the largest power decrease as the shading response benchmark point, monitors the phased changes, judges the shading evolution trend, and establishes shading status monitoring records.

[0121] The temperature and electrical consistency determination module calls the shading status monitoring record to obtain the temperature sampling sequence and power output sequence of the component during the monitoring period. It makes a corresponding judgment on the direction and magnitude of the change in each point in the temperature rise change sequence and power decay change sequence, analyzes the correspondence between the shading status and the temperature and electrical response, verifies the shading status, and obtains the shading response judgment result.

[0122] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for photovoltaic array shadowing condition detection, characterized in that, Includes the following steps: S1: Call the electrical measurement data of the photovoltaic array, analyze the open-circuit voltage and short-circuit current endpoints in the IV curve, calculate the available output range, determine the offset of the maximum power point on the voltage axis and current axis, and determine the degree of compression by comparing with the power point position under rated conditions, and obtain the compression offset strength coefficient. S2: Based on the compression offset strength coefficient, compare the voltage output of each series branch, sort them, calculate the difference between the current sort and the reference sort, construct the offset sequence, and determine the clustering and change direction in the physical arrangement order to obtain the voltage order disturbance distribution. S3: Using the voltage order disturbance distribution, calculate the power difference between adjacent branches and construct a power difference sequence. Analyze the changes in the difference in the sequence, identify the turning point of the power change trend, and number and mark the corresponding branches of the trend turning point to obtain the shielding level turning branch sequence number. S4: Call the shading level turning point branch number, locate the trend turning point branch and the power decrease change trajectory of adjacent deployment areas, extract the abrupt change node and select the node with the largest power decrease as the shading response reference point, monitor the phased changes, judge the shading evolution trend, and establish a shading status monitoring record.

2. The method for detecting the shading status of a photovoltaic array according to claim 1, characterized in that, The compression offset intensity coefficient includes voltage direction offset amplitude, current direction offset amplitude, and electrical working area coverage ratio. The voltage order disturbance distribution includes branch voltage sorting difference sequence, sequence offset cluster segment, and offset direction change characteristics. The shielding level turning branch sequence number is specifically the power difference change abrupt change position number, power change trend inflection point number, and branch physical layout corresponding number. The shielding status monitoring record includes power decrease segment change trajectory, abrupt change node amplitude sequence, and trend continuity time marker.

3. The method for detecting the shading status of a photovoltaic array according to claim 1, characterized in that, The specific steps for obtaining the compression offset strength coefficient are as follows: S111: Acquire electrical measurement data of photovoltaic array, collect open-circuit voltage endpoint and short-circuit current endpoint of IV curve, calculate the interval length in voltage direction and current direction, determine the length of available output interval based on endpoint difference, and generate available output interval value; S112: Based on the available output range value, determine the offset position of the maximum power point on the voltage axis and the current axis, call the maximum power point coordinates calibrated under the rated operating state, and compare the differences in the corresponding directions of the voltage axis and the current axis coordinates to obtain the maximum power point offset. S113: Based on the maximum power point offset, analyze the impact of the offset behavior on the coverage range of the available output interval, determine the degree of compression of the output interval coverage range by the offset behavior, construct the masking structure interference characteristics, and obtain the compression offset intensity coefficient.

4. The method for detecting the shading status of a photovoltaic array according to claim 3, characterized in that, The specific steps for obtaining the voltage order disturbance distribution are as follows: S211: Based on the compression offset strength coefficient, compare the voltage output values ​​of each branch of the photovoltaic array at the same time, sort the branches according to the output magnitude, and generate a branch voltage sorting sequence. S212: Call the branch voltage sorting sequence, calculate the position difference between the current sorting position and the preset reference position, record the sorting offset of each branch, and establish a voltage sorting offset sequence; S213: Based on the voltage sorting offset sequence, by performing continuity analysis, determine the aggregation characteristics and change direction of sorting differences in the physical arrangement order of branches, identify local disturbance sections of the sorting structure, and generate voltage order disturbance distribution.

5. The method for detecting the shading status of a photovoltaic array according to claim 4, characterized in that, The specific steps for obtaining the shading level transition branch sequence number are as follows: S311: Based on the voltage order disturbance distribution, call the power output values ​​of adjacent branches of the photovoltaic array, perform difference calculation on physically adjacent branches, record the power difference between each adjacent branch according to the deployment order, and generate a branch power difference sequence. S312: Call the branch power difference sequence, compare and calculate the adjacent differences in the sequence, analyze the increase change between adjacent differences, determine the abnormal position of the difference amplitude change in the power difference sequence, record the turning point of the power difference change, and obtain the turning point value of the power difference sequence. S313: Based on the power difference sequence turning point value, and based on the output change direction and amplitude characteristics of the branches on both sides of the power trend turning point, the corresponding branches of the trend turning point are numbered and marked to generate the shading level turning point branch sequence number.

6. The method for detecting the shading status of a photovoltaic array according to claim 5, characterized in that, The specific steps for obtaining the shading status monitoring record are as follows: S411: Based on the shielding level turning branch sequence number, call the power output continuous monitoring sequence, obtain the power output data of the trend turning branch and adjacent deployment areas, analyze the change trajectory of power output in the decreasing section, identify and locate the sudden node during the decreasing process, and obtain the sudden node location result. S412: Based on the mutation node location results, call the power output data corresponding to the mutation node, compare the power difference of each node in the descent segment, select the node with the largest power difference as the shading response reference point, calculate the power change amplitude and time difference of each sampling point relative to the reference point in the continuous monitoring sequence, construct the change rate sequence corresponding to the reference point, and use the formula: ； Calculate the influence value of the rate of change of the reference point to obtain the trend value of the reference power change; in, The rate of change influence value associated with the reference point. For the first The power normalization value of each sampling point The power normalization value at the reference point, For the first Normalized time values ​​for each sampling point Normalized time values ​​at the reference point This is the index number of the current sampling point in the continuous monitoring sequence. This represents the total number of sampling points in the current power monitoring section; S413: Call the reference power change trend value, analyze the change direction of the power difference sequence of the sampling points in the monitoring section, determine the stage structure between the power change direction and change amplitude in the continuous time segment, form the evolution trajectory of the shading state in the time dimension, and obtain the shading state monitoring record.

7. The method for detecting the shading status of a photovoltaic array according to claim 1, characterized in that, The method further includes: S5: Call the shading state monitoring record, obtain the temperature sampling sequence and power output sequence corresponding to the component during the monitoring period, make a corresponding judgment on the direction and magnitude of the change in each point in the temperature rise change sequence and power decay change sequence, analyze the correspondence of the shading state at the temperature and electrical response level, verify the shading state, and obtain the shading response judgment result. The shading response judgment results specifically refer to the thermal response trend sequence, the electrical response trend sequence, and the thermoelectric response correspondence analysis results.

8. The method for detecting the shading status of a photovoltaic array according to claim 7, characterized in that, The specific steps for obtaining the occlusion response judgment result are as follows: S511: Call the shading state monitoring record, obtain the temperature sampling sequence corresponding to the component during the monitoring period, perform difference judgment based on the temperature difference between adjacent sampling points in the sequence, construct the temperature rise change trajectory using the difference direction and difference amplitude, generate a change magnitude sequence based on the continuity of the trajectory in the time sequence, and obtain the temperature rise change amplitude. S512: Based on the temperature rise change magnitude, call the power output sequence of the component during the monitoring period, construct the power attenuation change trajectory based on the power difference between adjacent sampling points, analyze the change magnitude of the trajectory in the time sequence, make a correspondence judgment on the change direction and magnitude of each point in the temperature rise change sequence and the power attenuation change sequence, and calculate the response matching relationship value. S513: Based on the response matching relationship value, determine the correspondence of the shading state at the thermoelectric response level, verify the shading state, and obtain the shading response judgment result.

9. A system for detecting the shading status of a photovoltaic array, characterized in that, The system is used to implement the method for detecting the shading status of a photovoltaic array as described in any one of claims 1-8, the system comprising: The parameter range analysis module calls the electrical measurement data of the photovoltaic array, analyzes the open-circuit voltage and short-circuit current endpoints in the IV curve, calculates the available output range, determines the offset of the maximum power point on the voltage axis and current axis, and determines the degree of compression by comparing it with the power point position under rated conditions, and obtains the compression offset strength coefficient. The branch order sequencing module compares the voltage output of each branch according to the compression offset intensity coefficient, calculates the difference between the current sort and the reference sort after sorting, constructs the offset sequence, and determines the aggregation and change direction in the physical arrangement order to obtain the voltage order perturbation distribution. The power transition identification module uses the voltage order disturbance distribution to calculate the power difference between adjacent branches and construct a power difference sequence. It analyzes the changes in the difference in the sequence, identifies the turning point of the power change trend, and numbers and marks the corresponding branches of the trend turning point to obtain the shielding level turning branch sequence number. The shading evolution tracking module calls the shading level turning branch sequence number, locates the trend turning branch and the power decrease change trajectory of adjacent deployment areas, extracts abrupt change nodes and selects the node with the largest power decrease as the shading response benchmark point, monitors the stage changes, judges the shading evolution trend, and establishes a shading status monitoring record. The temperature and electrical consistency determination module calls the shading state monitoring record to obtain the temperature sampling sequence and power output sequence corresponding to the component during the monitoring period. It makes a corresponding judgment on the direction and magnitude of the change in each point in the temperature rise change sequence and power decay change sequence, analyzes the correspondence between the shading state and the temperature and electrical response, verifies the shading state, and obtains the shading response judgment result.

Images