Method for correcting output power of photovoltaic string combiner box

Abstract

A method for correcting the output power of a photovoltaic string combiner box comprises the following steps that a data set W of the output power of a photovoltaic string is obtained, and W = {W1, W2,..., Wn}; denoising the data set W to obtain a processed second data set; constructing a linear equation related to the output power of the combiner box according to the inclined surface irradiance of the photovoltaic string, giving an initial parameter to the linear equation, performing fitting according to the second data set, and updating the initial parameter to obtain a precise parameter; replacing the initial parameters with the accurate parameters to obtain a final equation; and obtaining the output power of the combiner box by obtaining the inclined surface irradiance of the photovoltaic string and the final equation. According to the method, the output power of the combiner box is directly obtained through the inclined plane irradiance of the photovoltaic string, and before the linear equation is obtained, the data set W is denoised to ensure that the final equation is correct. And accurate data is provided for each system.

Description

Technical Field

[0001] The invention relates to the technical field of combiner boxes, and in particular to a method for correcting the output power of a photovoltaic string combiner box. Background Art

[0002] The combiner box is an important component in a solar photovoltaic power station. It is mainly used to collect the DC current of multiple photovoltaic strings onto a centralized DC cable. The combiner box is usually equipped with a monitoring device to monitor the current, voltage, power and other parameters of the photovoltaic strings in real time, and transmit the monitoring data to the monitoring system so as to monitor and analyze the operation of the photovoltaic power station. Therefore, the accuracy of the monitoring data plays a vital role in the normal operation of the photovoltaic power station.

[0003] However, as the service time increases and components age, errors may occur in the combiner box data monitoring. When errors occur, cable connection problems, string failures, current imbalances, etc. cannot be fed back in time through current data, thereby delaying the time to solve the problem. Therefore, a method that can correct the output power of the positive photovoltaic string combiner box is urgently needed. Summary of the invention

[0004] In view of the above-mentioned defects, the purpose of the present invention is to propose a method for correcting the output power of a photovoltaic string combiner box, which directly obtains the output power of the combiner box through the irradiance of the inclined surface of the photovoltaic string, thereby avoiding the occurrence of monitoring data errors due to equipment aging.

[0005] To achieve this purpose, the present invention adopts the following technical solution: A method for outputting power of a positive photovoltaic string combiner box, comprising the following steps:

[0006] Step S1: Obtain a data set W of the output power of a photovoltaic string, where W = {W1, W2..., Wn};

[0007] Step S2: De-noise the data set W to obtain a processed second data set;

[0008] Step S3: constructing a linear equation related to the output power of the combiner box according to the irradiance of the inclined surface of the photovoltaic string, giving initial parameters of the linear equation, fitting according to the second data set, updating the initial parameters, and obtaining accurate parameters;

[0009] Use precise parameters to replace the initial parameters and get the final equation;

[0010] Step S4: Obtain the output power of the combiner box by obtaining the irradiance of the inclined surface of the photovoltaic string and the final equation.

[0011] Preferably, the specific steps of step S2 are as follows:

[0012] Step S21: randomly select t sample points from the data set W, and reconstruct a random sample set X, wherein the random sample set X contains multiple subsets;

[0013] Step S22: Select a subset X′ from the sample set X, and use the subset X′ as the root space of the tree;

[0014] Step S23: randomly generate a random point p in the subset X′;

[0015] Step S23: Divide the subset X′ space into two subspaces, and put sample points with dimensions less than p into the left subspace, and sample points with dimensions greater than or equal to p into the right subspace;

[0016] Step S24: randomly regenerate random points p in the left subspace and the right subspace respectively, and put the sample points with dimensions less than p into the left subspace, and put the sample points with dimensions greater than or equal to p into the right subspace;

[0017] Step S25: Repeat step S24 until there is only one sample point in the space, or the height of the tree reaches the height threshold, and obtain a processed tree;

[0018] Step S26: repeat steps S21 to S25 to obtain multiple processing trees;

[0019] Step S27: Select one of the sample points x i , let it traverse each processing tree and obtain its noise score value;

[0020] Determine whether the noise score is greater than the abnormal threshold. If so, delete the sample point. If not, keep the sample point.

[0021] Step S28: Repeat step S27 until all sample points are processed, and the remaining sample points constitute the second data set.

[0022] Preferably, the formula for obtaining the noise score value in step S27 is as follows:

[0023]

[0024] Where h(x) represents the number of divisions that the sample point x goes through from the root space of the tree to its leaf space, E() represents the expected value of each sample point, a is a threshold parameter, and c(a) represents the length of the path length.

[0025] in

[0026] H(a-1)=ln(a-1)+β;

[0027] Where β is a constant.

[0028] Preferably, the linear equation in step S3 is:

[0029] f(x) = dlnO + b;

[0030] Where d and b are initial parameters, and O is the irradiance of the inclined surface.

[0031] Preferably, the steps of fitting according to the second data set are as follows:

[0032] Set the step size and iteration threshold;

[0033] The linear equation is adjusted through a random search algorithm, and the value of the error function is obtained to determine whether the error function reaches the minimum value or the number of iterations reaches the threshold. If the error function reaches the minimum value or reaches the iteration threshold, the d value and b value of the current linear equation are output.

[0034] Preferably, the error function is as follows:

[0035] T = ∑(W′(x)-f(x)) 2 ;

[0036] Wherein W′(x) is the power value of the second data set.

[0037] One of the above technical solutions has the following advantages or beneficial effects: the present invention directly obtains the output power of the combiner box through the irradiance of the inclined surface of the photovoltaic string, and before obtaining the linear equation, the data set W is denoised to ensure that the final equation is correct. Accurate data is provided for each system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flow chart of an embodiment of the method of the present invention. DETAILED DESCRIPTION

[0039] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0040] In the description of the embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0041] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0042] like Figure 1 As shown, a method for outputting power of a photovoltaic string combiner box includes the following steps:

[0043] Step S1: Obtain a data set W of the output power of a photovoltaic string, where W = {W1, W2..., Wn};

[0044] Step S2: De-noise the data set W to obtain a processed second data set;

[0045] Step S3: constructing a linear equation related to the output power of the combiner box according to the irradiance of the inclined surface of the photovoltaic string, giving initial parameters of the linear equation, fitting according to the second data set, updating the initial parameters, and obtaining accurate parameters;

[0046] Use precise parameters to replace the initial parameters and get the final equation;

[0047] Step S4: Obtain the output power of the combiner box by obtaining the irradiance of the inclined surface of the photovoltaic string and the final equation.

[0048] Since the output power of the junction box will also change accordingly with the change of the irradiance of the inclined surface, there is a linear relationship between the irradiance of the inclined surface and the output power of the junction box in the actual working environment. Therefore, the output power of the junction box can be calculated by the irradiance of the inclined surface. At this time, the output power of the junction box can be directly calculated by the irradiance of the inclined surface to avoid the situation where the output power is wrong due to the aging of the components. However, when obtaining the linear relationship between the two, it is necessary to obtain the power output by the current photovoltaic string for calculation, but in the power data set, there will be data errors due to equipment aging. If these erroneous data are used to calculate the linear relationship, the final linear relationship is also wrong. Therefore, before obtaining the linear relationship between the two, it is necessary to denoise the data set W and remove discrete data (i.e., unreasonable data generated by equipment aging). For this reason, in the present invention, before obtaining the linear equation, it is necessary to execute step S2 to denoise the data set W to ensure that the final equation is correct. After obtaining the final equation, the output power of the junction box can be directly obtained by the irradiance of the inclined surface of the photovoltaic string, providing accurate data for each system.

[0049] Preferably, the specific steps of step S2 are as follows:

[0050] Step S21: randomly select t sample points from the data set W, and reconstruct a random sample set X, wherein the random sample set X contains multiple subsets;

[0051] Step S22: Select a subset X′ from the sample set X, and use the subset X′ as the root space of the tree;

[0052] Step S23: randomly generate a random point p in the subset X′;

[0053] Step S23: Divide the subset X′ space into two subspaces, and put sample points with dimensions less than p into the left subspace, and sample points with dimensions greater than or equal to p into the right subspace;

[0054] Step S24: randomly regenerate random points p in the left subspace and the right subspace respectively, and put the sample points with dimensions less than p into the left subspace, and put the sample points with dimensions greater than or equal to p into the right subspace;

[0055] Step S25: Repeat step S24 until there is only one sample point in the space, or the height of the tree reaches the height threshold, and obtain a processed tree;

[0056] Step S26: repeat steps S21 to S25 to obtain multiple processing trees;

[0057] Step S27: Select one of the sample points x i, let it traverse each processing tree and obtain its noise score value;

[0058] Determine whether the noise score is greater than the abnormal threshold. If so, delete the sample point. If not, keep the sample point.

[0059] Step S28: Repeat step S27 until all sample points are processed, and the remaining sample points constitute the second data set.

[0060] In the present invention, since a high-precision linear equation needs to be obtained, a large amount of data is required to iterate when fitting the equation to obtain the d value and the b value in the linear equation.

[0061] In order to ensure accuracy and improve efficiency, multiple sets of subsets X' are obtained by random sampling multiple times through the data set W, which can reduce the amount of calculation while ensuring accuracy. The subset X' is divided into two parts by random points p, which can ensure excellent processing efficiency when processing large-scale data. At the same time, multiple subsets X' are independent of each other and will not affect each other when constructing the processing tree, which can improve the sensitivity of outliers, so that the corresponding discrete points (noise points) can be detected in independent data subsets.

[0062] Preferably, the formula for obtaining the noise score value in step S27 is as follows:

[0063]

[0064] Where h(x) represents the number of divisions that the sample point x goes through from the root space of the tree to its leaf space, E() represents the expected value of each sample point, a is a threshold parameter, and c(a) represents the length of the path length.

[0065] in

[0066] H(a-1)=ln(a-1)+β;

[0067] Where β is a constant.

[0068] Preferably, the linear equation in step S3 is:

[0069] f(x) = dlnO + b;

[0070] Where d and b are initial parameters, and O is the irradiance of the inclined surface.

[0071] Preferably, the steps of fitting according to the second data set are as follows:

[0072] Set the step size and iteration threshold;

[0073] The linear equation is adjusted through a random search algorithm, and the value of the error function is obtained to determine whether the error function reaches the minimum value or the number of iterations reaches the threshold. If the error function reaches the minimum value or reaches the iteration threshold, the d value and b value of the current linear equation are output.

[0074] Preferably, the error function is as follows:

[0075] T=Σ(W′(x)-f(x)) 2 ;

[0076] Wherein W′(x) is the power value of the second data set.

[0077] In one embodiment of the present invention, the residual sum of squares of the fitting curve is calculated. The gradient and Hessian matrix of the parameters are calculated. The gradient represents the rate of change of the error function for each parameter, while the Hessian matrix represents the rate of change of the gradient. This information will be used to determine the adjustment direction and step size of the parameters. Using the information of the gradient and the Hessian matrix, the value of the parameter is adjusted by the Levenberg-Marquardt algorithm. The algorithm will consider the change of the current parameter and the change of the error function to determine the next moving direction and step size of the parameter. Repeat the steps until the stop condition is reached, the maximum iteration threshold reaches 1000 or the error function reaches the minimum value to determine the parameter values ​​corresponding to the d value and the b value.

[0078] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0079] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for outputting power of a positive photovoltaic string combiner box, characterized in that: The steps include: Step S1: Obtain a data set W of the output power of a photovoltaic string, where W = {W1, W2..., Wn}; Step S2: De-noise the data set W to obtain a processed second data set; Step S3: constructing a linear equation related to the output power of the combiner box according to the irradiance of the inclined surface of the photovoltaic string, giving initial parameters of the linear equation, fitting according to the second data set, updating the initial parameters, and obtaining accurate parameters; Use precise parameters to replace the initial parameters and get the final equation; Step S4: Obtain the output power of the combiner box by obtaining the irradiance of the inclined surface of the photovoltaic string and the final equation.

2. A method for outputting power of a positive photovoltaic string combiner box according to claim 1, characterized in that: The specific steps of step S2 are as follows: Step S21: randomly select t sample points from the data set W, and reconstruct a random sample set X, wherein the random sample set X contains multiple subsets; Step S22: Select a subset X′ from the sample set X, and use the subset X′ as the root space of the tree; Step S23: randomly generate a random point p in the subset X′; Step S23: Divide the subset X′ space into two subspaces, and put sample points with dimensions less than p into the left subspace, and sample points with dimensions greater than or equal to p into the right subspace; Step S24: randomly regenerate random points p in the left subspace and the right subspace respectively, and put the sample points with dimensions less than p into the left subspace, and put the sample points with dimensions greater than or equal to p into the right subspace; Step S25: Repeat step S24 until there is only one sample point in the space, or the height of the tree reaches the height threshold, and obtain a processed tree; Step S26: repeat steps S21 to S25 to obtain multiple processing trees; Step S27: Select one of the sample points x i , let it traverse each processing tree and obtain its noise score value; Determine whether the noise score is greater than the abnormal threshold. If so, delete the sample point. If not, keep the sample point. Step S28: Repeat step S27 until all sample points are processed, and the remaining sample points constitute the second data set.

3. A method for outputting power of a positive photovoltaic string combiner box according to claim 2, characterized in that: The formula for obtaining the noise score value in step S27 is as follows: Where h(x) represents the number of divisions that the sample point x goes through from the root space of the tree to its leaf space, E() represents the expected value of each sample point, a is a threshold parameter, and c(a) represents the length of the path length. in H(a-1)=ln(a-1)+β; Where β is a constant.

4. A method for outputting power of a positive photovoltaic string combiner box according to claim 1, characterized in that: The linear equation in step S3 is: f(x) = dlnO + b; Where d and b are initial parameters, and O is the irradiance of the inclined surface.

5. A method for outputting power of a positive photovoltaic string combiner box according to claim 4, characterized in that: The steps for fitting the second data set are as follows: Set the step size and iteration threshold; The linear equation is adjusted through a random search algorithm, and the value of the error function is obtained to determine whether the error function reaches the minimum value or the number of iterations reaches the threshold. If the error function reaches the minimum value or reaches the iteration threshold, the d value and b value of the current linear equation are output.

6. A method for outputting power of a positive photovoltaic string combiner box according to claim 5, characterized in that: The error function is as follows: T=Σ(W′(x)-f(x)) 2 ; Wherein W′(x) is the power value of the second data set.

Images