A method for detecting local shading and estimating the degree of mismatch of a photovoltaic array

Abstract

The present invention discloses a method for detecting local shadows and estimating the degree of mismatch in a photovoltaic array, comprising the steps of: (1) obtaining the I-V data of the actual operation of the photovoltaic array and performing preprocessing; (2) analyzing the I-V data of the actual operation of the photovoltaic array, and defining variables such as knee points, inflection points, light transmittance, local shadow detection threshold variables, and mismatch degree estimation threshold variables; (3) performing local shadow detection and mismatch degree estimation of the photovoltaic array, detecting shadows according to the local shadow detection threshold variables, and further selecting corresponding mismatch degree estimation processes according to the inflection points, light transmittance, and mismatch degree estimation threshold variables. The present invention detects local shadows based on the current-voltage data of the photovoltaic array, and further estimates the degree of mismatch, providing suggestions for the autonomous operation and maintenance of the photovoltaic system.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic system operation and maintenance, and particularly relates to a method for detecting local shadows of a photovoltaic array and estimating the degree of mismatch. Background Art

[0002] As an important part of a photovoltaic system, the operating state of a photovoltaic array directly determines the power generation efficiency and safety of the photovoltaic system. The photovoltaic array works in a harsh climate environment for a long time, and various faults are likely to occur, with many potential safety hazards. In particular, local shadows will cause hot spots on the photovoltaic modules, resulting in local temperature rise and even fires.

[0003] Technologies for detecting local shadows of a photovoltaic array are mainly divided into two categories. The first category is based on visible light images, which identify whether there are shadows on the surface of the photovoltaic array through image recognition. However, the cost of drones is relatively high, and the types of detected faults are single, which is not suitable for small and medium-sized power stations. The second category is based on the I-V curve of the photovoltaic array, which realizes local shadow detection through threshold analysis, machine learning, etc., and can detect multiple faults simultaneously, and is gradually being commercialized. However, the current technology can only achieve binary judgment, detecting whether there are local shadows in the photovoltaic array, and it is difficult to estimate the number of blocked modules and the degree of mismatch, which has limited reference value for the autonomous operation and maintenance of the photovoltaic system.

[0004] Therefore, it is necessary to detect local shadows of the photovoltaic array and estimate the degree of mismatch to provide suggestions for the autonomous operation and maintenance of the photovoltaic system. Summary of the Invention

[0005] Object of the Invention: Aiming at the above problems, the present invention proposes a method for detecting local shadows of a photovoltaic array and estimating the degree of mismatch. This method performs local shadow detection based on the current-voltage data of the photovoltaic array and further estimates the degree of mismatch.

[0006] Technical Solution: To achieve the object of the present invention, the technical solution adopted by the present invention is: A method for detecting local shadows of a photovoltaic array and estimating the degree of mismatch, specifically including the following steps:

[0007] Step 1, obtain the current-voltage curve data of the actual operation of the photovoltaic array and perform preprocessing to obtain the preprocessed current-voltage curve data;

[0008] Step 2, perform data analysis on the preprocessed current-voltage curve data in Step 1 to obtain the open-circuit point voltage, short-circuit point current, knee point, inflection point, and transmittance of the current-voltage curve data;

[0009] Step 3: Detect the local shadow of the photovoltaic array based on the data analysis results obtained in Step 2 to obtain the local shadow detection result of the photovoltaic array. If there is local shadow in the photovoltaic array, estimate the mismatch degree of the photovoltaic array according to the data analysis results obtained in Step 2; otherwise, end the step.

[0010] Further, the method of Step 1 is specifically as follows:

[0011] Step 1.1: Use an inverter with an integrated current-voltage data scanning function to obtain the current-voltage curve data of the actual operation of the photovoltaic array, denoted as [V ori , I ori ; where V ori is the voltage of each point of the current-voltage curve before preprocessing, and I ori is the current of each point of the current-voltage curve before preprocessing.

[0012] Step 1.2: Perform data preprocessing on the current-voltage curve data described in Step 1.1 to obtain the preprocessed current-voltage data, denoted as [V mea , I mea ; where V mea is the voltage of each point of the current-voltage curve after preprocessing, and I mea is the current of each point of the current-voltage curve after preprocessing.

[0013] Further, the preprocessing described in Step 1.2 includes, but is not limited to, outlier detection, linear interpolation, and smoothing filtering.

[0014] Further, the method of Step 2 is specifically as follows:

[0015] Step 2.1: Obtain the open-circuit point voltage, short-circuit point current, and power data of the actual operation of the photovoltaic array, specifically as follows:

[0016] V oc = max(V mea )

[0017] I sc = max(I mea )

[0018] P mea = I mea · V mea

[0019] In the formula, V oc represents the open-circuit point voltage of the actual operation of the photovoltaic array, I sc represents the short-circuit point current of the actual operation of the photovoltaic array, and P mea represents the power of the actual operation of the photovoltaic array.

[0020] Step 2.2, perform third-order differentiation on the current data at each point of the actual operation of the photovoltaic array, and obtain the maximum and minimum values of the second-order derivative of the current data at each point, as follows:

[0021]

[0022] maxI d2 =max(I d2 )

[0023] minI d2 =min(I d2 )

[0024] In the formula, I d1 represents the first-order derivative of the current data at each point of the actual operation of the photovoltaic array, I d2 represents the second-order derivative of the current data at each point of the actual operation of the photovoltaic array, I d3 represents the third-order derivative of the current data at each point of the actual operation of the photovoltaic array, maxI d2 represents the maximum value of the second-order derivative of the current data at each point of the actual operation of the photovoltaic array, minI d2 represents the minimum value of the second-order derivative of the current data at each point of the actual operation of the photovoltaic array;

[0025] Step 2.3, perform first-order differentiation on the power-voltage data of the actual operation of the photovoltaic array, and obtain the knee point according to the first-order differentiation result. The knee point is the local maximum power point, as follows:

[0026]

[0027] K(j)={i|P d1 (i)>0∩P d1 (i + 1)<0}

[0028] In the formula, P d1 represents the first-order derivative of the power-voltage data of the actual operation of the photovoltaic array, K(j) represents the jth knee point, P d1 (i) represents the first-order derivative of the power-voltage data of the actual operation of the ith photovoltaic array, P d1 (i + 1) represents the first-order derivative of the power-voltage data of the actual operation of the (i + 1)th photovoltaic array;

[0029] Step 2.4, obtain the inflection point of the current-voltage curve data of the actual operation of the photovoltaic array, as follows:

[0030] S(k)={i|I d3 (i)·I d3 (i + 1)<0∩I d3 (i - 1)>0∩I d3(i + 2) < 0 ∩ I d2 (i) > Threshold_max}

[0031] Wherein, S(k) represents the k-th inflection point of the current-voltage curve data of the actual operation of the photovoltaic array, and I d3 (i) represents the third derivative of the current-voltage data of the i-th photovoltaic array in actual operation, and I d3 (i + 1) represents the third derivative of the current-voltage data of the (i + 1)-th photovoltaic array in actual operation, and I d3 (i - 1) represents the third derivative of the current-voltage data of the (i - 1)-th photovoltaic array in actual operation, and I d3 (i + 2) represents the third derivative of the current-voltage data of the (i + 2)-th photovoltaic array in actual operation, and I d2 (i) represents the second derivative of the current-voltage data of the i-th photovoltaic array in actual operation; Threshold_max represents the maximum value of the local shadow detection threshold variable;

[0032] Step 2.5, according to the current at the inflection point and the short-circuit point current of the actual operation of the photovoltaic array, calculate the light transmittance at this inflection point, specifically as follows:

[0033]

[0034] Wherein, I_s(k) represents the current at the inflection point S(k), and Tr(k) represents the light transmittance at the inflection point S(k).

[0035] Furthermore, in the said step 3, according to the data analysis result obtained in step 2, perform local shadow detection on the photovoltaic array, and the specific method is as follows:

[0036] Compare the open-circuit point voltage of the actual operation of the photovoltaic array with the preset local shadow detection threshold to obtain the local shadow detection result of the photovoltaic array, including:

[0037]

[0038] Wherein, PS_state represents the local shadow detection result of the photovoltaic array. PS_state = 1 indicates that there is local shadow in the photovoltaic array; PS_state = 0 indicates that there is no local shadow in the photovoltaic array; maxI d2 represents the maximum value of the second derivative of the current data of each point of the actual operation of the photovoltaic array, Threshold_max represents the maximum value of the local shadow detection threshold variable, and minI d2 represents the minimum value of the second derivative of the current data of each point of the actual operation of the photovoltaic array, and Threshold_min represents the minimum value of the local shadow detection threshold variable.

[0039] Further, the estimation of the mismatch degree of the photovoltaic array is performed according to the data analysis result obtained in step 3, and the specific method is as follows:

[0040] Judge the number K of all the inflection points S(k) obtained in step 2.4;

[0041] If K = 1 and the transmittance Tr(k) at this inflection point S(k), k = 1 satisfies Tr(k) < Threshold_ps,k = 1, then the estimation of the mismatch degree of the photovoltaic array is performed according to process A; Threshold_ps represents the preset mismatch degree estimation threshold variable;

[0042] If K = 1 and the transmittance Tr(k) at this inflection point S(k), k = 1 satisfies Tr(k) ≥ Threshold_ps,k = 1, then the estimation of the mismatch degree of the photovoltaic array is performed according to process B;

[0043] Otherwise, if K > 1, then the estimation of the mismatch degree of the photovoltaic array is performed according to process B.

[0044] Further, the specific method of the process A is as follows:

[0045] A1. Construct a photovoltaic array model and define the photovoltaic array model parameters as Sim_par = [I ph , I s , R s , R sh , a]; The specific photovoltaic array model is as follows:

[0046]

[0047] n = Num·N cells

[0048] R s = R s_cell ·n

[0049] R sh = R sh_cell ·n

[0050] In the formula, V is the output voltage of the photovoltaic array, I is the output current of the photovoltaic array, I ph is the photocurrent, I s is the reverse saturation current of the diode, R s is the equivalent series resistance of the photovoltaic array, R sh is the equivalent parallel resistance of the photovoltaic array, a is the ideal factor of the diode, q is the Coulomb charge, g is the Boltzmann constant, T is the Kelvin temperature, n is the number of photovoltaic array cells, Num is the number of photovoltaic modules, N cells is the number of cells in a single photovoltaic module, R s_cellis the equivalent series resistance of a single cell, R sh_cell is the equivalent parallel resistance of a single cell;

[0051] A2. Move the inflection point S(k), k = 1 on the current-voltage curve data 10 points to the low-voltage side to obtain the shifted inflection point S'(k), k = 1; and intercept the shifted inflection point S'(k), k = 1 and all the current-voltage data on its low-voltage side from the current-voltage curve data [V mea , I mea , and denote it as represents the voltage of each point of the intercepted current-voltage curve, represents the current of each point of the intercepted current-voltage curve;

[0052] where S'(k) represents the inflection point obtained by moving the k-th inflection point S(k) of the current-voltage curve data of the photovoltaic array in actual operation 10 points to the low-voltage side, k = 1;

[0053] A3. Use the metaheuristic optimization algorithm to extract the photovoltaic array model parameters Sim_par from the current-voltage curve segment ;

[0054] A4. Substitute the photovoltaic array model parameters Sim_par described in step A3 into the photovoltaic array model described in step A1 to obtain the reconstructed complete current-voltage curve data, denoted as and record the maximum voltage in the reconstructed complete current-voltage curve data as V oc_sim ; represents the voltage of each point of the reconstructed complete current-voltage curve, represents the current of each point of the reconstructed complete current-voltage curve;

[0055] A5. Calculate the number of shaded components N ps and the shading rate Rate according to the following formula:

[0056]

[0057]

[0058] Rate = 1 - Tr(k)

[0059] In the formula, round(·) is to retain one decimal place, V 0 is the voltage of a single bypass diode.

[0060] Furthermore, the specific method of the process B is as follows:

[0061] B1, construct the photovoltaic array model, and define the photovoltaic array model parameters as Sim_par = [I ph ,I s ,R s ,R sh ,a]; the photovoltaic array model is as follows:

[0062]

[0063] n=Num·N cells

[0064] R s =R s_cell ·n

[0065] R sh =R sh_cell ·n

[0066] Where V is the output voltage of the photovoltaic array, I is the output current of the photovoltaic array, and I ph is the photocurrent, I s is the diode reverse saturation current, R s is the photovoltaic array equivalent series resistance, R sh is the equivalent parallel resistance of the photovoltaic array, a is the diode ideal factor, q is the Coulomb charge, g is the Boltzmann constant, T is the Kelvin temperature, n is the number of photovoltaic array cells, Num is the number of photovoltaic modules, N cells is the number of cells in a single photovoltaic module, R s_cell is the equivalent series resistance of a single cell, R sh_cell is the equivalent parallel resistance of a single battery cell;

[0067] B2, intercept all the current-voltage curve data in the current-voltage curve segment of the first knee point and its low voltage side, recorded as And intercept all the current-voltage curve data in the current-voltage curve segment of the last knee point and its high voltage side, recorded as The and Perform splicing to obtain the spliced ​​current-voltage curve data set

[0068] B3, using meta-heuristic optimization algorithm to optimize the current-voltage data set Extract photovoltaic array model parameters Sim_par;

[0069] B4, substitute the photovoltaic array model parameter Sim_par described in step B3 into the photovoltaic array model described in step B1 to obtain the reconstructed complete current-voltage curve data, recorded as

[0070] B5. Use numerical methods to inversely calculate the number of photovoltaic modules N corresponding to [I_s(k), V_s(k)] at the inflection point k ; where V_s(k) represents the voltage at the inflection point S(k);

[0071] B6. Calculate the number of shaded modules N psk and the shading rate Rate k :

[0072]

[0073]

[0074] In the formula, N k+1 represents the number of photovoltaic modules corresponding to [I_s(k + 1), V_s(k + 1)] at the inflection point, and N end represents the number of photovoltaic modules corresponding to [I_s(end), V_s(end)] at the inflection point, and end represents the last point in the current-voltage curve data.

[0075] Furthermore, in step B5, when using numerical methods to inversely calculate the number of photovoltaic modules N corresponding to [I_s(k), V_s(k)] at the inflection point, the numerical methods include but are not limited to Newton's iteration method, bisection method, and Stephenson accelerated iteration method.

[0076] Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0077] The present invention performs local shadow detection based on the current-voltage data of the photovoltaic array, further estimates the degree of mismatch, and provides suggestions for the autonomous operation and maintenance of the photovoltaic system;

[0078] The present invention details the data analysis process and defines the knee point, local shadow detection threshold variable, inflection point, mismatch degree estimation threshold variable, and light transmittance.

[0079] The present invention details the process of estimating the degree of mismatch and realizes the estimation of the degree of mismatch under different light transmittances of the photovoltaic array. Description of the Drawings

[0080] Figure 1 is a flowchart of the technical solution described in the present invention under an embodiment;

[0081] Figure 2 is a flowchart of the mismatch degree estimation method described in the present invention under an embodiment;

[0082] Figure 3 is a sub-flowchart A of the mismatch degree estimation method described in the present invention under an embodiment;

[0083] Figure 4 It is a flowchart of sub - process B of the mismatch degree estimation method according to the present invention in an embodiment;

[0084] Figure 5 It is a schematic diagram of voltage - current curve data according to the present invention in an embodiment. Detailed implementation manners

[0085] The technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0086] A method for detecting local shadow and estimating mismatch degree of a photovoltaic array according to the present invention is characterized in that, referring to Figure 1 , it includes the following steps:

[0087] (1) Obtain the current - voltage data of the actual operation of the photovoltaic array and perform pre - processing;

[0088] Further, the step (1) specifically includes:

[0089] (1.1) Through an inverter integrated with the function of scanning current - voltage data, obtain the current - voltage data of the actual operation of the photovoltaic array, denoted as [V ori , I ori ;

[0090] (1.2) Perform data pre - processing on [V ori , I ori , ], mainly including outlier detection, linear interpolation, smoothing filtering, etc. The pre - processed current - voltage data is denoted as [V mea , I mea ;

[0091] (2) Perform data analysis on the current - voltage data of the actual operation of the photovoltaic array;

[0092] Further, referring to the attached Figure 5 figure, the step (2) specifically includes:

[0093] (2.1) Obtain the open - circuit point voltage, short - circuit point current and power data of the actual operation of the photovoltaic array, denoted as V oc , I sc and P mea , as shown in Equation (1);

[0094] V oc = max(V mea )

[0095] I sc = max(I mea ) (1)

[0096] P mea = I mea ·Vmea

[0097] (2.2) Third-order derivatives are respectively taken for the current data of the photovoltaic array, denoted as I d1 , I d2 , I d3 , as shown in Equation (2); the maximum and minimum values of the second-order derivative of the current are obtained, denoted as maxId2 and minId2 respectively, as shown in Equation (3);

[0098]

[0099]

[0100] (2.3) Third-order derivatives are respectively taken for the power data of the photovoltaic array, denoted as P d1 , as shown in Equation (4);

[0101]

[0102] (2.4) Obtain the position of the local maximum power point, defined as the knee point, denoted as K(j), as shown in Equation (5);

[0103] K(j) = {i | P d1 (i) > 0 ∩ P d1 (i + 1) < 0} (5)

[0104] In the formula, i is the i-th data point and j is the j-th knee point.

[0105] (2.5) Combining prior knowledge, define the local shadow detection threshold variables, denoted as Threshold_max and Threshold_min;

[0106] (2.6) Obtain the position of the depression in the current-voltage curve representation, defined as the inflection point, denoted as S(k), as shown in Equation (6) S(k) = {i | I d3 (i) · I d3 (i + 1) < 0 ∩ I d3 (i - 1) > 0 ∩ I d3 (i + 2) < 0 ∩ I d2 (i) > Threshold_max} (6)

[0107] In the formula, k is the k-th inflection point.

[0108] (2.7) Combining prior knowledge, define the mismatch degree estimation threshold variable, denoted as Threshold_ps;

[0109] (2.8) Denote the current and voltage at the inflection point S(k) as I_s(k) and V_s(k) respectively, and denote I_s(k) and I scThe ratio is defined as the transmittance, denoted as Tr(k), as shown in Equation (7);

[0110]

[0111] (3) Perform local shadow detection and mismatch degree estimation of the photovoltaic array;

[0112] Furthermore, the step (3) specifically includes:

[0113] (3.1) Perform local shadow detection of the photovoltaic array;

[0114] Furthermore, this step specifically means comparing the open - circuit point voltage with the local shadow detection threshold to obtain the local shadow detection result of the photovoltaic array, denoted as PS_state, as shown in Equation (8); if PS_state is 1, it indicates that there is local shadow in the photovoltaic array; if PS_state is 0, it indicates that there is no local shadow in the photovoltaic array.

[0115]

[0116] (3.2) Perform mismatch degree estimation of the photovoltaic array;

[0117] Furthermore, as shown in the appendix Figure 2 This step includes:

[0118] When PS_state = 1, according to the total number of inflection points K in the step (2.6) and the transmittance Tr(k) in the step (2.8), select different mismatch degree estimation processes;

[0119] (3.2.1) If K = 1, then when Tr(k) < Threshold_ps, estimate according to process A; when Tr(k) ≥ Threshold_ps, estimate according to process B;

[0120] (3.2.2) If K > 1, estimate according to process B;

[0121] Furthermore, as shown in the appendix Figure 3 This process A specifically includes:

[0122] A1 Move the position of the inflection point S(1) on the I - V curve 10 points to the low - voltage side to obtain a new inflection point S'(1), intercept the current data of the points after moving 10 points, denoted as Intercept the voltage data of the points after moving 10 points, denoted as i cut is the i cut th point of the intercepted curve segment, that is, the 1st to the S'(1)th point on the curve as shown in Equation (9):

[0123]

[0124] A2 Use the meta - heuristic optimization algorithm to extract the photovoltaic array model parameters, defined as the reconstruction parameter Sim_par;

[0125] Furthermore, the meta - heuristic optimization algorithm is the particle swarm optimization algorithm, and its objective function f is as follows. Sim_par makes the f value minimum:

[0126]

[0127] A3 Substitute the reconstruction parameter Sim_par into the above - mentioned photovoltaic array equivalent model to obtain the reconstructed current - voltage data, denoted as Obtain the maximum value of the reconstructed de voltage, denoted as V oc_sim ;

[0128] A4 Calculate the number of shaded - occluded components and the shading rate according to Equation (11), denoted as N ps and Rate respectively;

[0129]

[0130] In the formula, round(·) is to retain one decimal place, and V 0 is the voltage of a single bypass diode.

[0131] Furthermore, referring to Figure 4 , the process B specifically includes:

[0132] B1 Respectively intercept the current data on the low - voltage side of the first knee point, denoted as I cut1 , intercept the current data on the high - voltage side of the last knee point, denoted as I cut2 , splice the two segments of current data, denoted as k cut is the k cut th data point intercepted, K(1) is the first knee point among the knee points K(j), K end is the last knee point among the knee points K(j), and I end is the last data point in I mea , specifically as follows:

[0133]

[0134] B2 Respectively intercept the voltage data before the first knee point, denoted as V cut1 , intercept the voltage data after the last knee point, denoted as V cut2 , splice the two segments of voltage data, denoted as V end is Vmea The last point is as follows:

[0135]

[0136] B3 Utilize a meta-heuristic optimization algorithm to extract the photovoltaic array model parameters, defined as the reconstruction parameter Sim_par;

[0137] B4 Substitute the reconstruction parameter Sim_par into the above photovoltaic array equivalent model to obtain the reconstructed current-voltage data, denoted as

[0138] B5 Use a numerical method to inversely solve for the number of photovoltaic modules corresponding to [I_s(k), V_s(k)] at the inflection point, denoted as N k ;

[0139] Furthermore, in this embodiment, the Newton iteration method is selected to solve for N k , and the target equation f is as follows:

[0140]

[0141] Take the first derivative of N(k), as follows:

[0142]

[0143] Iteratively solve according to Equation (18) until f is less than 0.01 to obtain the approximate solution N k

[0144]

[0145] B6 Calculate the number of shaded components N psk and the shading rate Rate k :

[0146]

[0147] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations (such as knee points, inflection points, numerical methods, mismatch degree estimation processes, etc.) can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for detecting local shading and estimating mismatch degree of a photovoltaic array, characterized in that, it specifically includes the following steps: Step 1, obtain the current-voltage curve data of the actual operation of the photovoltaic array and perform preprocessing to obtain the preprocessed current-voltage curve data; Step 2, perform data analysis on the preprocessed current-voltage curve data obtained in Step 1 to obtain the open-circuit point voltage, short-circuit point current, knee point, inflection point, and light transmittance of the preprocessed current-voltage curve data; Step 3, according to the data analysis results obtained in Step 2, detect the local shading of the photovoltaic array to obtain the local shading detection result of the photovoltaic array; if there is local shading in the photovoltaic array, estimate the mismatch degree of the photovoltaic array according to the data analysis results obtained in Step 2; otherwise, end Step 3; In Step 3, according to the data analysis results obtained in Step 2, perform local shading detection of the photovoltaic array. The specific method is as follows: Compare the open-circuit point voltage of the actual operation of the photovoltaic array with the preset local shading detection threshold to obtain the local shading detection result of the photovoltaic array, including: Wherein, PS_state represents the local shadow detection result of the photovoltaic array. When PS_state = 1, it indicates that there is local shadow in the photovoltaic array; when PS_state = 0, it indicates that there is no local shadow in the photovoltaic array; maxI d2 represents the maximum value of the second derivative of the current data of each point in the actual operation of the photovoltaic array, Threshold_max represents the maximum value of the local shadow detection threshold variable, minI d2 represents the minimum value of the second derivative of the current data of each point in the actual operation of the photovoltaic array, Threshold_min represents the minimum value of the local shadow detection threshold variable; The method for estimating the mismatch degree of the photovoltaic array according to the data analysis results obtained in Step 2 in Step 3 is as follows: Judge the number K of all the inflection points S(k) obtained in Step 2.4; If K = 1, and when the light transmittance Tr(k) at this inflection point S(k), k = 1 < Threshold_ps, k = 1 holds, then estimate the mismatch degree of the photovoltaic array according to Process A; Threshold_ps represents the preset mismatch degree estimation threshold variable; If K = 1, and when the light transmittance Tr(k) at this inflection point S(k), k = 1 ≥ Threshold_ps, k = 1 holds, then estimate the mismatch degree of the photovoltaic array according to Process B; Otherwise, when K > 1, then estimate the mismatch degree of the photovoltaic array according to Process B.

2. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 1, characterized in that, The method of Step 1 is as follows: Step 1.1, use an inverter with integrated current-voltage data scanning function to obtain the current-voltage curve data of the actual operation of the photovoltaic array, denoted as [V ori , I ori ; where, V ori is the voltage of each point on the current-voltage curve, and I ori is the current of each point on the current-voltage curve; Step 1.2, perform data preprocessing on the current-voltage curve data described in Step 1.1 to obtain the preprocessed current-voltage curve data, denoted as [V mea , I mea ; where V mea is the voltage of each point on the preprocessed current-voltage curve, and I mea is the current of each point on the preprocessed current-voltage curve.

3. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 1 or 2, characterized in that, The preprocessing includes, but is not limited to, abnormal point detection, linear interpolation, and smoothing filtering.

4. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 1, characterized in that, The method of Step 2 is as follows: Step 2.1, obtain the open-circuit point voltage, short-circuit point current, and power data of the actual operation of the photovoltaic array, specifically as follows: V oc = max(V mea ) I sc = max(I mea ) P mea = I mea · V mea Where, V oc represents the open-circuit point voltage of the photovoltaic array during actual operation, I sc represents the short-circuit point current of the photovoltaic array during actual operation, P mea represents the power of the photovoltaic array during actual operation; V mea is the voltage at each point of the pre-processed current-voltage curve, I mea is the current at each point of the pre-processed current-voltage curve; Step 2.2, perform third-order derivative on the current data of each point of the actual operation of the photovoltaic array respectively, and obtain the maximum and minimum values of the second-order derivative of the current data of each point, specifically as follows: maxI d2 = max(I d2 ) minI d2 = min(I d2 ) Wherein, I d1 represents the first derivative of the current data of each point in the actual operation of the photovoltaic array, I d2 represents the second derivative of the current data of each point in the actual operation of the photovoltaic array, I d3 represents the third derivative of the current data of each point in the actual operation of the photovoltaic array, maxI d2 represents the maximum value of the second derivative of the current data of each point in the actual operation of the photovoltaic array, minI d2 represents the minimum value of the second derivative of the current data of each point in the actual operation of the photovoltaic array; Step 2.3, perform first-order derivative on the power-voltage data of the actual operation of the photovoltaic array, and obtain the knee point according to the first-order derivative result. The knee point is the local maximum power point, specifically as follows: K(j) = {i | P d1 (i) > 0 ∩ P d1 (i + 1) < 0} Wherein, P d1 represents the first derivative of the power-voltage data of the actual operation of the photovoltaic array, K(j) represents the j-th knee point, and P d1 (i) represents the first derivative of the power-voltage data of the actual operation of the i-th photovoltaic array, and P d1 (i + 1) represents the first derivative of the power-voltage data of the actual operation of the (i + 1)-th photovoltaic array; Step 2.4, obtain the inflection point of the current-voltage curve data of the actual operation of the photovoltaic array, specifically as follows: S(k) = {i | I d3 (i) · I d3 (i + 1) < 0 ∩ I d3 (i - 1) > 0 ∩ I d3 (i + 2) < 0 ∩ I d2 (i) > Threshold_max} Wherein, S(k) represents the k-th inflection point of the current-voltage curve data of the actual operation of the photovoltaic array, and I d3 (i) represents the third derivative of the current-voltage data of the actual operation of the i-th photovoltaic array, and I d3 (i + 1) represents the third derivative of the current-voltage data of the actual operation of the (i + 1)-th photovoltaic array, and I d3 (i - 1) represents the third derivative of the current-voltage data of the actual operation of the (i - 1)-th photovoltaic array, and I d3 (i + 2) represents the third derivative of the current-voltage data of the actual operation of the (i + 2)-th photovoltaic array, and I d2 (i) represents the second derivative of the current-voltage data of the actual operation of the i-th photovoltaic array; Threshold_max represents the maximum value of the local shadow detection threshold variable; Step 2.5, calculate the transmittance at the inflection point according to the current at the inflection point and the short - circuit point current of the actual operation of the photovoltaic array, specifically as follows: In the formula, \(I_s(k)\) represents the current at the inflection point \(S(k)\), and \(Tr(k)\) represents the transmittance at the inflection point \(S(k)\).

5. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 4, characterized in that the specific method of the process A is as follows: A1. Build a photovoltaic array model and define the parameters of the photovoltaic array model as Sim_par = [I ph , I s , R s , R sh , a]; The specific photovoltaic array model is as follows: n = Num·N cells R s = R s_cell ·n R sh = R sh_cell ·n Wherein, V is the output voltage of the photovoltaic array, I is the output current of the photovoltaic array, and I ph is the photocurrent, I s is the reverse saturation current of the diode, R s is the equivalent series resistance of the photovoltaic array, R sh is the equivalent parallel resistance of the photovoltaic array, a is the ideality factor of the diode, q is the Coulomb charge, g is the Boltzmann constant, T is the Kelvin temperature, n is the number of photovoltaic array cells, Num is the number of photovoltaic modules, N cells is the number of cells in a single photovoltaic module, R s_cell is the equivalent series resistance of a single cell, R sh_cell is the equivalent parallel resistance of a single cell; A2, shift the upper inflection point S(k), k = 1 of the current-voltage curve data 10 points towards the low-voltage side to obtain the shifted inflection point S'(k), k = 1; and intercept the shifted inflection point S'(k), k = 1 and all the current-voltage data on its low-voltage side from the current-voltage curve data [V mea , I mea , and denote it as represents the voltage of each point of the intercepted current-voltage curve, represents the current of each point of the intercepted current-voltage curve; wherein, \(S'(k)\) represents the inflection point obtained by moving the position of the \(k\) - th inflection point \(S(k)\) of the current - voltage curve data of the actual operation of the photovoltaic array 10 points to the low - voltage side, and \(k = 1\); A3, using a metaheuristic optimization algorithm for the current-voltage curve segment Extract the photovoltaic array model parameters Sim_par; A4. Substitute the photovoltaic array model parameters Sim_par described in step A3 into the photovoltaic array model described in step A1 to obtain the reconstructed complete current-voltage curve data, denoted as and record that the maximum voltage in the reconstructed complete current-voltage curve data is V oc_sim ; represents the voltage of each point on the reconstructed complete current-voltage curve, represents the current of each point on the reconstructed complete current-voltage curve; A5, calculate the number of shadow occlusion components N according to the following formula ps and the light shielding rate Rate: Rate=1 - Tr(k) where round(·) is to retain one decimal place, and V 0 is the voltage of a single bypass diode.

6. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 5, characterized in that the specific method of the process B is as follows: B1, construct a photovoltaic array model and define the parameters of the photovoltaic array model as Sim_par = [I ph , I s , R s , R sh , a]; The specific photovoltaic array model is as follows: n = Num·N cells R s = R s_cell ·n R sh = R sh_cell ·n Where, V is the output voltage of the photovoltaic array, I is the output current of the photovoltaic array, I ph is the photocurrent, I s is the reverse saturation current of the diode, R s is the equivalent series resistance of the photovoltaic array, R sh is the equivalent parallel resistance of the photovoltaic array, a is the ideality factor of the diode, q is the Coulomb charge, g is the Boltzmann constant, T is the Kelvin temperature, n is the number of photovoltaic array cells, Num is the number of photovoltaic modules, N cells is the number of cells in a single photovoltaic module, R s_cell is the equivalent series resistance of a single cell, R sh_cell is the equivalent parallel resistance of a single cell; For B2, intercept all the current-voltage curve data in the current-voltage curve segment on the low-voltage side of the first knee point and denote it as and intercept all the current-voltage curve data in the current-voltage curve segment on the high-voltage side of the last knee point and denote it as Combine the and to obtain the combined set of current-voltage curve data represents the voltage of each point on the combined current-voltage curve, represents the current of each point on the combined current-voltage curve; B3. Using a metaheuristic optimization algorithm for the current-voltage data set Extract the photovoltaic array model parameters Sim_par; B4. Substitute the photovoltaic array model parameters Sim_par described in step B3 into the photovoltaic array model described in step B1 to obtain the reconstructed complete current-voltage curve data, denoted as B5. Using a numerical method to inversely calculate the number of photovoltaic modules N corresponding to [I_s(k), V_s(k)] at the inflection point k ; where I_s(k) represents the current at the inflection point S(k), and V_s(k) represents the voltage at the inflection point S(k). B6. Calculate the number N of shadow occlusion components according to the following formula psk and the light shielding rate Rate k : where N k+1 represents the number of photovoltaic modules corresponding to [I_s(k + 1), V_s(k + 1)] at the inflection point, and N end represents the number of photovoltaic modules corresponding to [I_s(end), V_s(end)] at the inflection point, and end represents the last point in the current-voltage curve data.

7. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 6, characterized in that in step B5, the numerical method for inversely obtaining the number of photovoltaic modules corresponding to \([I_s(k),V_s(k)]\) at the inflection point, and the numerical method includes but is not limited to Newton's iteration method, bisection method and Stephenson accelerated iteration method.

8. A method for detecting local shading and estimating mismatch degree of a photovoltaic array according to claim 5 or 6, characterized in that the meta - heuristic optimization algorithm is specifically a particle swarm optimization algorithm, and the objective function formula is: In the formula, \(f\) represents the objective function value of the meta - heuristic optimization algorithm.

Images