Improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance

Abstract

The invention discloses a BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance improvement, and belongs to the field of photovoltaic power station energy efficiency evaluation, and the method specifically comprises the steps: obtaining the related data of photovoltaic power station monitoring equipment, and calculating the theoretical output power P0 of a photovoltaic array; a photovoltaic power station system is divided into a photovoltaic array power generation module, a combiner box transmission module, an inverter conversion transmission module and an alternating current grid-connected transmission module, the efficiency of the four modules is calculated respectively, and a photovoltaic power station comprehensive evaluation index is established; calculating the weight of the comprehensive evaluation index of the photovoltaic power station through an entropy weight method and a CRITIC method, and carrying out combined weighting by combining with an improved minimum identification information principle to obtain the combined weight of the comprehensive evaluation index of the photovoltaic power station; and performing energy efficiency evaluation on the photovoltaic power station by using a BMW-TOPSIS evaluation model improved by an Euclidean-KL distance, thereby realizing abnormality diagnosis of the power generation state of the photovoltaic power station.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power station energy efficiency evaluation, and in particular relates to a photovoltaic power station energy efficiency evaluation method based on an improved BMW-TOPSIS Euclidean-KL distance. Background Art

[0002] As global demand for clean energy continues to grow, photovoltaic power plants, as a key component of renewable energy, are rapidly expanding in both speed and scale. However, PV power plants face numerous operational challenges, such as aging equipment, dust pollution, and environmental factors. These factors can lead to reduced power generation efficiency, impacting their economic benefits and sustainable development. Therefore, energy efficiency evaluation of PV power plants is particularly important. This not only provides a strong basis for operation and maintenance, but also effectively improves their overall performance and economic benefits.

[0003] Currently, PV power plant energy efficiency assessment methods primarily include statistical analysis and machine learning. These methods primarily focus on analyzing the efficiency of a single device or component, lacking a comprehensive assessment of the energy efficiency of the entire PV power plant system. This makes it difficult to determine the operating status of key PV plant equipment and the overall performance of the plant. When power generation efficiency issues arise in a PV plant, it's difficult to quickly pinpoint the problematic link. Furthermore, some existing assessment methods are complex, requiring large amounts of data and tedious calculation steps. This makes them impractical in practical applications and makes them difficult to meet the demand for rapid and accurate PV power plant assessments.

[0004] As mentioned in the review, the existing photovoltaic power station energy efficiency evaluation method is unable to determine the operating status of important equipment in the photovoltaic power station and the process is complicated, and it cannot quickly reflect the actual power generation status of the photovoltaic power station. Summary of the Invention

[0005] To overcome the shortcomings of the existing technology, the present invention aims to provide a photovoltaic power plant energy efficiency evaluation method based on an improved BMW-TOPSIS method based on the Euclidean-KL distance. The specific technical solutions of the present invention are as follows:

[0006] The BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on the Euclidean-KL distance improvement includes the following steps:

[0007] Step 1: Obtain relevant data from the photovoltaic power station monitoring equipment and calculate the theoretical output power P0 of the photovoltaic array using the photovoltaic cell single diode equivalent circuit model;

[0008] Step 2: Divide the PV power station system into four parts: the PV array power generation module, the combiner box transmission module, the inverter conversion transmission module, and the AC grid-connected transmission module. Calculate the efficiency of each of these four modules. Establish a comprehensive evaluation index for the PV power station based on the acquired PV power station monitoring equipment data and the calculated efficiencies of the four modules.

[0009] Step 3: Calculate the weights of the comprehensive evaluation indicators of the photovoltaic power station using the entropy weight method and the CRITIC method, and combine them with the improved minimum discriminant information principle to obtain the combined weights of the comprehensive evaluation indicators of the photovoltaic power station;

[0010] Step 4: Use the BMW-TOPSIS evaluation model improved by Euclidean-KL distance to evaluate the energy efficiency of the photovoltaic power station, thereby realizing abnormal diagnosis of the power generation status of the photovoltaic power station.

[0011] Furthermore, in step 2, the photovoltaic array power generation module includes a photovoltaic array part, the combiner box transmission module includes a DC line from the photovoltaic array to the combiner box and a combiner box part, the inverter conversion transmission module includes a DC line from the combiner box to the inverter and an inverter part, and the AC grid-connected transmission module includes an AC line part from the inverter output end to the grid connection point; the comprehensive evaluation index of the photovoltaic power station is the calculated efficiency of the four parts: the photovoltaic array power generation module efficiency n1, the combiner box transmission module efficiency n2, the inverter conversion transmission module efficiency n3, and the AC grid-connected transmission module efficiency n4; wherein, the photovoltaic array power generation module efficiency is The efficiency of the combiner box transmission module is The efficiency of the inverter conversion transmission module is The efficiency of AC grid-connected transmission module is P1 is the output power of the photovoltaic array, P2 is the output power of the combiner box, P3 is the output power of the inverter, and P4 is the grid connection point power.

[0012] Furthermore, in step 3, the entropy weight method measures the importance of each indicator by the entropy value of the standardized data of the photovoltaic power station comprehensive evaluation index, and calculates the weight vector α of each indicator by using it. j ,…,α n ], the calculation method is as follows:

[0013]

[0014] Where N ij is the normalized value of the jth indicator in the i-th time series of the photovoltaic power station, f ij is the proportion of the value of the jth indicator in the i-th time series to the sum of the jth indicators in all time series data, e j is the information entropy of the jth indicator, α jis the weight of indicator j under the entropy weight method.

[0015] Furthermore, in step 3, the CRITIC method determines the weight vector β of each indicator by calculating the differences and conflicts of the comprehensive evaluation indicators of the photovoltaic power station. j ,…,β n ], the calculation method is as follows:

[0016]

[0017] Where s j is the standard deviation of the j-th indicator, is the mean of the jth indicator, λ jk is the Pearson correlation coefficient between the jth indicator and the kth indicator, H j The conflict of the jth indicator, β j is the weight of indicator j under the CRITIC method.

[0018] Furthermore, in step 3, the index weight vector calculated by the entropy weight method and the CRITIC method is used to determine the combined weight vector ω=(ω1,ω2,…,ω j ,…,ω n ), the calculation formula is as follows:

[0019]

[0020] The calculation formula of the combined weight vector is:

[0021]

[0022] Where w j is the combined weight of index j, R t represents the information carrying capacity of the t-th weighting method, σ t represents the variance of the weights of each indicator in the tth weighting method, and ρ represents the Spearman correlation coefficient between the two weighting methods.

[0023] Furthermore, the step 4 specifically includes:

[0024] Step 4.1: Construct a standardized weighted evaluation matrix C based on the comprehensive evaluation index of the photovoltaic power station in step 2 and the combined weight vector in step 3. m×n ;

[0025] Step 4.2, determine the positive ideal solution B, the median ideal solution M, and the negative ideal solution W, and based on the weak dominance principle, all c ij It is divided into two categories: BM and MW, and the BM category is superior to the MW category;

[0026] Step 4.3, calculate c in BM class ij The Euclidean-KL distance between the positive ideal solution B and the median ideal solution M and the c in the MW class ij The Euclidean-KL distance between the median ideal solution M and the negative ideal solution W;

[0027] Step 4.4, calculate the progress of the BM and MW segments;

[0028] In step 4.5, the calculated progress is sorted, and the 20th and 80th percentiles of the progress are used as two thresholds to classify the PV power station into three energy efficiency evaluation levels: healthy, sub-healthy, and abnormal. An early warning is issued for abnormal conditions.

[0029] Furthermore, in step 4.1, the weighted evaluation matrix C m×n The calculation formula is:

[0030] C m×n =(C ij ) m×n =(N ij w j ) m×n ,

[0031] Where w j is the combined weight of index j, C ij Standardized weighted evaluation matrix C m×n The values of each element in .

[0032] Furthermore, in step 4.3, the Euclidean distance formula is:

[0033]

[0034] KL distance formula:

[0035]

[0036] Where B j is the maximum value of the j-th index of matrix B; M j is the median of the j-th index of matrix M; W j is the minimum value of the j-th index of matrix W;

[0037] Euclidean-KL distance formula:

[0038]

[0039] Where λ is the preference for the Euclidean distance measure. The preference for the two distance measures is the same, so λ = 0.5.

[0040] Furthermore, in step 4.4, the calculation formula for posting progress is:

[0041]

[0042] Where, T i The progress of the post, the value range is [0,1].

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] The method of this application makes up for the deficiency of single Euclidean distance in reflecting relative position relationships, makes the ranking more convincing, and effectively improves the accuracy of photovoltaic power station energy efficiency evaluation; by dividing the photovoltaic power station into four modules and constructing a comprehensive evaluation index based on this, the calculation is simplified, and the operating status of key equipment such as photovoltaic modules and inverters can be accurately judged; and based on the improved minimum discriminant information principle, the indicator weights obtained by the entropy weight method and CRITIC are combined and weighted, which solves the problem of unreasonable indicator weighting caused by the average distribution of weight coefficients in the commonly used combined weighting method, making the calculation results more reasonable and scientific. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of the evaluation method of the present invention;

[0046] Figure 2 This is a workflow diagram of the photovoltaic power station energy efficiency evaluation model of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] like Figure 1 As shown, a photovoltaic power station energy efficiency evaluation method based on the improved BMW-TOPSIS Euclidean-KL distance includes the following steps:

[0049] (I) Step 1: Obtain relevant data from the photovoltaic power station monitoring equipment, including: irradiance, temperature, photovoltaic array output power P1, combiner box output power P2, inverter output power P3, grid connection point power P4, and other operating data for one year at an interval of 15 minutes. Perform data preprocessing such as cleaning and noise reduction, and calculate the theoretical output power P0 of the photovoltaic array using the photovoltaic cell single diode equivalent circuit model.

[0050] (2) Step 2: Divide the photovoltaic power station system into four parts: photovoltaic array power generation module, combiner box transmission module, inverter conversion transmission module, and AC grid-connected transmission module. Calculate the efficiency of these four modules separately, and establish a comprehensive evaluation index for the photovoltaic power station based on the acquired photovoltaic power station monitoring equipment data and the calculated efficiency of the four modules.

[0051] The photovoltaic array power generation module includes the photovoltaic array part, the combiner box transmission module includes the DC line from the photovoltaic array to the combiner box and the combiner box part, the inverter conversion transmission module includes the DC line from the combiner box to the inverter and the inverter part, and the AC grid-connected transmission module includes the AC line part from the inverter output end to the grid connection point.

[0052] The comprehensive evaluation index is the calculated efficiency of the photovoltaic array power generation module n1, the combiner box transmission module efficiency n2, the inverter conversion transmission module efficiency n3, and the AC grid-connected transmission module efficiency n4. Among them, the photovoltaic array power generation module efficiency is The efficiency of the combiner box transmission module is The efficiency of the inverter conversion transmission module is The efficiency of AC grid-connected transmission module is

[0053] (III) Step 3: Calculate the weights of the comprehensive evaluation indicators of the photovoltaic power station by using the entropy weight method and the CRITIC method, and combine them with the improved minimum discriminant information principle to obtain the combined weights of the comprehensive evaluation indicators of the photovoltaic power station.

[0054] The entropy weight method measures the importance of each indicator by the entropy value of the standardized data of the comprehensive evaluation indicators of the photovoltaic power station, and calculates the weight vector of each indicator α=[α1,α2,…,α j ,…,α n ]; CRITIC method determines the weight vector of each index by calculating the difference and conflict of comprehensive evaluation indexes of photovoltaic power station β=[β1,β2,…,β j ,…,β n ].

[0055] The entropy weight method is calculated as follows:

[0056] The original data of comprehensive evaluation index of photovoltaic power station is normalized, and the value range is [0,1].

[0057]

[0058] Where, X ij is the original data of the jth indicator in the i-th time series of the photovoltaic power station, X max and X min is the maximum and minimum value of the indicator, N ij is the normalized value of the jth indicator in the i-th time series of the photovoltaic power station;

[0059]

[0060] Where, f ijis the proportion of the value of the jth indicator in the i-th time series to the sum of the jth indicators in all time series data, e j is the information entropy of the jth indicator, α j is the weight of indicator j under the entropy weight method.

[0061] The CRITIC method is calculated as follows:

[0062]

[0063] Where s j is the standard deviation of the j-th indicator, is the mean of the jth indicator, λ jk is the Pearson correlation coefficient between the jth indicator and the kth indicator, H j is the conflict of the jth indicator, β j is the weight of indicator j under the CRITIC method.

[0064] The indicator weight vectors calculated by the above two methods are used to determine the combined weight vector ω=(ω1,ω2,…,ω j ,…,ω n ), the calculation method is as follows:

[0065]

[0066] The calculation formula of the combined weight vector is:

[0067]

[0068] Where w j is the combined weight of index j, R t represents the information carrying capacity of the t-th weighting method, σ t represents the variance of the weights of each indicator in the tth weighting method, and ρ represents the Spearman correlation coefficient between the two weighting methods.

[0069] (IV) Step 4: Use the BMW-TOPSIS evaluation model improved by the Euclidean-KL distance to evaluate the energy efficiency of the photovoltaic power station, thereby realizing abnormal diagnosis of the power generation status of the photovoltaic power station.

[0070] like Figure 2 As shown, the specific process of this step is as follows:

[0071] ① Based on the comprehensive evaluation index of the photovoltaic power station in step 2 and the combined weight vector in step 3, a standardized weighted evaluation matrix C is constructed. m×n :

[0072] C m×n =(C ij )m×n =(N ij w j ) m×n ,

[0073] Where w j is the combined weight of index j. ij Standardized weighted evaluation matrix C m×n The values of each element in .

[0074] ② Determine the positive ideal solution B, the median ideal solution M, and the negative ideal solution W, and based on the weak dominance principle, all c ij It is divided into two categories: BM and MW, and the BM category is superior to the MW category;

[0075]

[0076] In the formula, the symbol ≥ means "better than", and the symbol # means all j c ij Better than M j The number of j For reference point M in attribute c ij The weighted normalized evaluation value under .

[0077] ③Calculate c in BM class ij The Euclidean-KL distance between the positive ideal solution B and the median ideal solution M and the c in the MW class ij The Euclidean-KL distance between the median ideal solution M and the negative ideal solution W.

[0078] Euclidean distance formula:

[0079]

[0080]

[0081] KL distance formula:

[0082]

[0083] Where B j is the maximum value of the j-th index of matrix B; M j is the median of the j-th index of matrix M; W j is the minimum value of the j-th index of matrix W.

[0084] Euclidean-KL distance formula:

[0085]

[0086] λ is the preference for Euclidean distance measure. The preference for the two distance measures is the same, so λ = 0.5 is taken.

[0087] ④Calculate the progress of the BM and MW segments. The calculation formula is:

[0088]

[0089] Where, T i The progress of the post, the value range is [0,1].

[0090] ⑤ Based on the calculated progress of the installation, the 20th and 80th percentiles of the installation progress are taken as two thresholds to divide the photovoltaic power station into three energy efficiency assessment levels: healthy, sub-healthy and abnormal, and issue early warnings for abnormal conditions.

[0091] The evaluation model inputs are the comprehensive evaluation indicators for the PV power plant obtained in step 2 and the weight vector for the indicator combination determined in step 3. The output is the energy efficiency rating of the PV power plant. If the evaluation result is abnormal, the efficiency of the four modules of the PV power plant system is checked to accurately determine the operating status of key operating equipment, providing a scientific basis for subsequent operation, maintenance, and management.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance is characterized by: The following steps are involved: Step 1: Obtain relevant data from the photovoltaic power station monitoring equipment and calculate the theoretical output power P0 of the photovoltaic array using the photovoltaic cell single diode equivalent circuit model; Step 2: Divide the PV power station system into four parts: the PV array power generation module, the combiner box transmission module, the inverter conversion transmission module, and the AC grid-connected transmission module. Calculate the efficiency of each of these four modules. Establish a comprehensive evaluation index for the PV power station based on the acquired PV power station monitoring equipment data and the calculated efficiencies of the four modules. Step 3: Calculate the weights of the comprehensive evaluation indicators of the photovoltaic power station using the entropy weight method and the CRITIC method, and combine them with the improved minimum discriminant information principle to obtain the combined weights of the comprehensive evaluation indicators of the photovoltaic power station; Step 4: Use the BMW-TOPSIS evaluation model improved by Euclidean-KL distance to evaluate the energy efficiency of the photovoltaic power station, thereby realizing abnormal diagnosis of the power generation status of the photovoltaic power station.

2. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 1 is characterized in that: In step 2, the photovoltaic array power generation module includes a photovoltaic array part, the combiner box transmission module includes a DC line from the photovoltaic array to the combiner box and the combiner box part, the inverter conversion transmission module includes a DC line from the combiner box to the inverter and the inverter part, and the AC grid-connected transmission module includes an AC line part from the inverter output end to the grid connection point; the comprehensive evaluation index of the photovoltaic power station is the calculated efficiency of the four parts: photovoltaic array power generation module efficiency n1, combiner box transmission module efficiency n2, inverter conversion transmission module efficiency n3, and AC grid-connected transmission module efficiency n4; wherein, the photovoltaic array power generation module efficiency is The efficiency of the combiner box transmission module is The efficiency of the inverter conversion transmission module is The efficiency of AC grid-connected transmission module is P1 is the output power of the photovoltaic array, P2 is the output power of the combiner box, P3 is the output power of the inverter, and P4 is the grid connection point power.

3. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 1 is characterized in that: In step 3, the entropy weight method measures the importance of each indicator by the entropy value of the standardized data of the photovoltaic power station comprehensive evaluation index, and calculates the weight vector α of each indicator by using it. j ,…,α n ], the calculation method is as follows: Where N ij is the normalized value of the jth indicator in the i-th time series of the photovoltaic power station, f ij is the proportion of the value of the jth indicator in the i-th time series to the sum of the jth indicators in all time series data, e j is the information entropy of the jth indicator, α j is the weight of indicator j under the entropy weight method.

4. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 3 is characterized in that: In step 3, the CRITIC method determines the weight vector β of each indicator by calculating the differences and conflicts of the comprehensive evaluation indicators of the photovoltaic power station. j ,…,β n ], the calculation method is as follows: Where s j is the standard deviation of the j-th indicator, is the mean of the jth indicator, λ jk is the Pearson correlation coefficient between the jth indicator and the kth indicator, H j The conflict of the jth indicator, β j is the weight of indicator j under the CRITIC method.

5. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 4 is characterized in that: In step 3, the index weight vector calculated by the entropy weight method and the CRITIC method is used to determine the combined weight vector ω=(ω1,ω2,…,ω j ,…,ω n ), the calculation formula is as follows: The calculation formula of the combined weight vector is: Where w j is the combined weight of index j, R t represents the information carrying capacity of the t-th weighting method, σ t represents the variance of the weights of each indicator in the tth weighting method, and ρ represents the Spearman correlation coefficient between the two weighting methods.

6. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 5 is characterized in that: The step 4 specifically includes: Step 4.1: Based on the comprehensive evaluation index of the photovoltaic power station in step 2 and the combined weight vector in step 3, a standardized weighted evaluation matrix C is constructed. m×n ; Step 4.2, determine the positive ideal solution B, the median ideal solution M, and the negative ideal solution W, and based on the weak dominance principle, all c ij It is divided into two categories: BM and MW, and the BM category is superior to the MW category; Step 4.3, calculate c in BM class ij The Euclidean-KL distance between the positive ideal solution B and the median ideal solution M and the c in the MW class ij The Euclidean-KL distance between the median ideal solution M and the negative ideal solution W; Step 4.4, calculate the progress of the BM and MW segments; In step 4.5, the calculated progress is sorted, and the 20th and 80th percentiles of the progress are used as two thresholds to classify the PV power station into three energy efficiency evaluation levels: healthy, sub-healthy, and abnormal. An early warning is issued for abnormal conditions.

7. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 6 is characterized in that: In step 4.1, the weighted evaluation matrix C m×n The calculation formula is: C m×n =(C ij ) m×n =(N ij w j ) m×n , Where w j is the combined weight of index j, C ij Standardized weighted evaluation matrix C m×n The values of each element in .

8. The improved BMW-TOPSIS photovoltaic power station energy efficiency evaluation method based on Euclidean-KL distance according to claim 6 is characterized in that: In step 4.3, the Euclidean distance formula is: KL distance formula: Where B j is the maximum value of the j-th index of matrix B; M j is the median of the j-th index of matrix M; W j is the minimum value of the j-th index of matrix W; Euclidean-KL distance formula: Where λ is the preference for the Euclidean distance measure. The preference for the two distance measures is the same, so λ = 0.

5.

9. The method for evaluating photovoltaic power station energy efficiency based on the improved BMW-TOPSIS Euclidean-KL distance according to claim 6, characterized in that: In step 4.4, the calculation formula for the posting progress is: Where, T i The progress of the post, the value range is [0,1].

Images