A method for evaluating the confidence of errors between field test indexes and theoretical analysis

Abstract

This invention provides a method for assessing the confidence level of field test indicators and theoretical analysis errors, relating to the field of advanced energy technology. The method includes: identifying key meteorological parameters by analyzing the sources of uncertainty in measurement data; establishing a comprehensive comparison relationship and mapping function for measurement data by analyzing the probability distribution of measurement data under key meteorological parameters; calculating the standard uncertainty under different evaluation results based on electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function; calculating the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty; calculating the environmental correction coefficient of the measurement data using the effective degrees of freedom of the combined standard uncertainty and a table lookup based on the confidence level; and applying the calculated environmental correction coefficient to the measurement data to adjust the measurement data. This invention ensures dynamic compatibility between the uncertainty and accuracy of field measurement results.

Description

Technical Field

[0001] This invention relates to the field of advanced energy technology, and more specifically, to a method for assessing the confidence level of errors in field testing indicators and theoretical analysis. Background Technology

[0002] Because on-site testing of photovoltaic power plants is greatly affected by environmental factors such as temperature, and the interaction and coupling mechanisms between environmental and operational factors are complex, they have a significant impact on the operating parameters of testing instruments and various components of the photovoltaic system under test. This leads to limitations in existing laboratory standard environments (STC, i.e., solar irradiance of 1000 W / m²). 2 The measurement data and error evaluation methods (with battery temperature at 25℃ and atmospheric mass at 1.5) are not applicable to error analysis of field data and quality judgment of field measurement parameters. There is no data accumulation of deviation compensation methods and error correction coefficients for environmental impact that can be used as a reference, which makes it impossible to accurately evaluate the actual output parameters of photovoltaic power plants. Therefore, it is of great significance to propose a field test index and theoretical analysis error confidence assessment method based on comparative statistical analysis.

[0003] Currently, there is no research in China on the evaluation system and methods for the confidence level of on-site testing indicators and theoretical analysis errors of photovoltaic power plants. Only inventive methods for the output performance of photovoltaic power plants have been proposed. For example, Chinese patent CN101441239B discloses a method for verifying the power generation performance of a grid-connected photovoltaic power plant. First, the solar irradiance of the solar photovoltaic module's light-receiving surface and the temperature of the photovoltaic cells inside the photovoltaic module are collected at intervals. Combined with the total power of the photovoltaic power plant modules, the power irradiance coefficient and power temperature coefficient of the solar photovoltaic modules, as well as the DC correction coefficient, inverter efficiency, and attenuation coefficient of the solar photovoltaic modules, the instantaneous power generation is calculated according to the formula. The instantaneous power generation is then multiplied by the time interval of the data collection and summed to obtain the theoretical power generation over a period of time. This is then compared with the actual power generation data to verify whether the power generation performance of the grid-connected photovoltaic power plant meets the requirements of the contract, design, or regulations.

[0004] Chinese patent CN101728984B discloses a method for predicting the power generation of a grid-connected photovoltaic (PV) power plant. This method utilizes input parameters (including solar radiation intensity, ambient temperature, and wind speed) and output parameters (i.e., output power) collected at the PV power plant production site to establish a database linking these input and output parameters. This database is then updated in real-time through online self-learning. For a given prediction of the input parameters of a grid-connected PV power plant, data mining techniques are used to mine data from the database linking these parameters, yielding a predicted value for the power generation of the grid-connected PV power plant.

[0005] None of the aforementioned patents elaborated on the error confidence level, and there are currently no relevant patents that provide an explanation or description of it.

[0006] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0007] In view of this, the present invention provides a method for evaluating the confidence level of field test indicators and theoretical analysis errors to solve the aforementioned problems.

[0008] To solve the above problems, the specific technical solution adopted by the present invention is as follows:

[0009] A method for evaluating the confidence level of errors between field test indicators and theoretical analysis includes:

[0010] S1. Collect meteorological environmental data and electrical quantity measurement data of photovoltaic power station respectively, and screen out key meteorological parameters by analyzing the source of uncertainty of measurement data. By analyzing the probability distribution law of measurement data under key meteorological parameters, establish a comprehensive comparison relationship and mapping function of measurement data.

[0011] S2. Based on the electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the standard uncertainty under different evaluation results, and calculate the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty;

[0012] S3. Utilize the effective degrees of freedom of the combined standard uncertainty, and calculate the environmental correction coefficient of the measurement data by looking up the table in conjunction with the confidence level. Apply the calculated environmental correction coefficient to the measurement data to adjust the measurement data.

[0013] Preferably, the process of separately collecting meteorological environmental data and electrical quantity measurement data of the photovoltaic power station, filtering out key meteorological parameters by analyzing the sources of uncertainty in the measurement data, and establishing a comprehensive comparison relationship and mapping function for the measurement data by analyzing the probability distribution law of the measurement data under the key meteorological parameters includes:

[0014] S11. Based on the meteorological environment data and electrical quantity measurement data of the photovoltaic power station, select the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions.

[0015] S12. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, analyze the sources of uncertainty in the measurement data using the niche theory, and screen out key meteorological parameters using correlation analysis.

[0016] S13. Select measurement data under different key meteorological parameters, analyze the probability distribution law of the measurement data under key meteorological parameters through data fitting technology, and establish a comprehensive comparison relationship and mapping function of the measurement data based on the probability distribution law.

[0017] Preferably, the step of analyzing the sources of uncertainty in the measurement data using niche theory based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, and selecting key meteorological parameters using correlation analysis includes:

[0018] S121. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, the final weights of meteorological environmental parameters are determined using the fuzzy MACBETH method.

[0019] S122. For each meteorological environmental parameter, calculate the niche overlap between measurement data under different conditions and measurement data under the same conditions, based on the weight of the meteorological environmental parameter.

[0020] S123. Using the Pearson correlation coefficient method, calculate the correlation coefficient between the measurement data and various meteorological environmental parameters to obtain the correlation analysis results.

[0021] S124. Based on the results of niche overlap and correlation analysis, comprehensively evaluate the impact of each meteorological parameter on the uncertainty of measurement data, and determine the key meteorological parameters.

[0022] Preferably, the step of determining the final weights of meteorological environmental parameters using the fuzzy MACBETH method based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions includes:

[0023] S1211. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, the semantic scale of the fuzzy MACBETH method represents the decision-maker's subjective judgment on the importance of different meteorological environmental parameters.

[0024] S1212. For each decision-maker, construct a fuzzy comparison matrix of meteorological and environmental parameters based on the semantic scale.

[0025] S1213. Construct a fuzzy linear programming model with the objective function of minimizing the sum of squares of the differences in the weights of all meteorological and environmental parameters.

[0026] S1214. By solving the fuzzy linear programming model, the fuzzy weights of each meteorological environmental parameter are obtained, and the fuzzy weights are normalized to obtain the final weights of the meteorological environmental parameters.

[0027] Preferably, the step of calculating the niche overlap between measurement data under different conditions and measurement data under the same conditions, based on the weight of each meteorological environmental parameter, includes:

[0028] S1221. Standardize the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions respectively.

[0029] S1222. For each meteorological environmental parameter, the measurement data under the same conditions shall be regarded as the ecological niche space.

[0030] S1223. For measurement data under different conditions, the degree of overlap between the measurement data under the same conditions and the measurement data under the same conditions in the niche space is calculated using cosine similarity to obtain the niche overlap degree.

[0031] Preferably, the step of calculating the standard uncertainty under different evaluation results based on electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, and calculating the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty includes:

[0032] S21. Based on the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions, calculate the standard uncertainty of the measurement data under the same meteorological conditions and the standard uncertainty of the measurement data under different single meteorological conditions respectively.

[0033] S22. Based on the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions, and combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the combined standard uncertainty.

[0034] S23. Based on the combined standard uncertainty, calculate the effective degrees of freedom of the combined standard uncertainty using the formula for calculating effective degrees of freedom.

[0035] Preferably, the formulas for calculating the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions are as follows:

[0036]

[0037] In the formula, u A (x) represents the standard uncertainty of measurement data under the same meteorological conditions, X represents the average value of measurement data under the same meteorological conditions, n-1 represents the degrees of freedom, and n represents the number of measurement data under the same meteorological conditions. i u represents the i-th measurement data in the measurement data under the same meteorological conditions. B (x) represents the standard uncertainty of measurement data under different conditions and a single meteorological environment, a represents the half width of the possible value interval of the measurand, and k represents the coverage factor.

[0038] Preferably, the formula for calculating the combined standard uncertainty is:

[0039]

[0040] In the formula, u c (y) represents the combined standard uncertainty, u(x) i ) represents the input quantity x i The standard uncertainty, f, represents the comprehensive comparison relationship and mapping function of the measurement data, u(x) i ,x j ) represents the input quantity x i With input quantity x j The covariance, where N represents the total number of measurement data inputs, x j Represents the relationship with input quantity x i The j-th unrelated measurement data, where x represents the measurement data.

[0041] Preferably, the formula for calculating the effective degrees of freedom is:

[0042]

[0043] In the formula, γ eff The effective degrees of freedom, u, represent the combined standard uncertainty. c (y) represents the combined standard uncertainty, u i (y) represents the standard uncertainty component of the i-th input quantity, γ i Let represent the degrees of freedom of the i-th measurement data input, N represent the total number of measurement data inputs, and i represent the ordinal number of the measurement data input.

[0044] Preferably, the step of using the effective degrees of freedom of the combined standard uncertainty, combined with confidence level lookup table calculation of the environmental correction coefficient for the measurement data, and applying the calculated environmental correction coefficient to the measurement data to adjust the measurement data includes:

[0045] S31. Determine the expansion factor by looking up the table based on the effective degrees of freedom and confidence level of the combined standard uncertainty;

[0046] S32. Calculate the expanded uncertainty of the measurement data based on the expansion factor and the combined standard uncertainty;

[0047] The formula for calculating the expanded uncertainty is as follows:

[0048] U=k′u c (y);

[0049] In the formula, U represents the expanded uncertainty, u c (y) represents the combined standard uncertainty, and k′ represents the expansion factor;

[0050] S33. Calculate the environmental correction factor for the testing instrument based on the expanded uncertainty;

[0051] S34. Apply the calculated environmental correction factor to the original measurement data to adjust the measurement data.

[0052] The beneficial effects of this invention are as follows:

[0053] This invention, based on thorough theoretical analysis and cross-verification with standard laboratory results, refines and improves models and parameters using advanced numerical processing and analysis techniques, establishes confidence assessment criteria, and proposes a confidence assessment method for the field test indicators of photovoltaic power plants and the errors in theoretical analysis based on comparative statistical analysis. This method ensures the dynamic compatibility of uncertainty and data accuracy in field measurement results; it also effectively assesses field measurement errors, guaranteeing the reliability of photovoltaic power plant output parameters; and by establishing environmental correction coefficients for measuring instruments, it proposes an evaluation method for the measurement uncertainty introduced by field measuring instruments in photovoltaic power plants. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0055] Figure 1 This is a flowchart of a method for evaluating the confidence level of field test indicators and theoretical analysis errors according to an embodiment of the present invention;

[0056] Figure 2 This is a flowchart illustrating the specific implementation of a method for evaluating the confidence level of field test indicators and theoretical analysis errors according to an embodiment of the present invention. Detailed Implementation

[0057] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0058] According to an embodiment of the present invention, a method for evaluating the confidence level of field test indicators and theoretical analysis errors is provided.

[0059] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1-2 As shown, the method for evaluating the confidence level of field test indicators and theoretical analysis errors according to an embodiment of the present invention includes:

[0060] S1. Collect meteorological environmental data and electrical quantity measurement data of photovoltaic power station respectively, and screen out key meteorological parameters by analyzing the source of uncertainty of measurement data. By analyzing the probability distribution law of measurement data under key meteorological parameters, establish a comprehensive comparison relationship and mapping function of measurement data.

[0061] In a preferred embodiment, the process of separately collecting meteorological environmental data and electrical quantity measurement data of the photovoltaic power station, filtering out key meteorological parameters by analyzing the sources of uncertainty in the measurement data, and establishing a comprehensive comparison relationship and mapping function for the measurement data by analyzing the probability distribution law of the measurement data under the key meteorological parameters includes:

[0062] S11. Based on the meteorological environment data and electrical quantity measurement data of the photovoltaic power station, select the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions.

[0063] It should be noted that typical geographic and meteorological environmental data are collected in real time, including solar irradiance (SI), ambient temperature (T), humidity (H), air pressure (P), direct radiation (DR), diffuse radiation (SR), ultraviolet radiation (UV), and wind speed / direction (W).

[0064] The testing instruments are determined as follows: IV curve analyzer, voltage sensor, current sensor, and the electrical quantities to be measured: battery module output voltage and current, inverter output voltage and current. Data on these electrical quantities is collected and generated into data sets {x1(SI,T,H,P,DR,SR,UV,W,),x2(SI,T,H,P,DR,SR,UV,W,)…x n (SI,T,H,P,DR,SR,UV,W,)}, the collected data is modeled, and two cases are divided:

[0065] (1) Select meteorological conditions under which repeated measurements are conducted, where the changes in meteorological parameters are very small, to form a measurement data model {x1,x2,…,x}. n ,};

[0066] (2) Repeated measurement data collection is performed under a single meteorological environment with different conditions. The influence of a certain meteorological environmental parameter on the measuring instrument is considered, and the sources of uncertainty of the testing instrument are analyzed and studied to form a measurement data model. For example, data can be selected under different SI environments, different T environments, and similarly, data can be selected under different H / P / DR / SR / UV / W environments; thus forming a measurement data model {x1(θ), x2(θ), ..., x n (θ)}, where θ is one of the meteorological environments mentioned above.

[0067] S12. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, analyze the sources of uncertainty in the measurement data using the niche theory, and screen out key meteorological parameters using correlation analysis.

[0068] In a preferred embodiment, the step of analyzing the sources of uncertainty in the measurement data using niche theory based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, and selecting key meteorological parameters using correlation analysis includes:

[0069] S121. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, the final weights of meteorological environmental parameters are determined using the fuzzy MACBETH method.

[0070] In a preferred embodiment, the step of determining the final weights of meteorological environmental parameters using the fuzzy MACBETH method based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions includes:

[0071] S1211. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, the semantic scale of the fuzzy MACBETH method represents the decision-maker's subjective judgment on the importance of different meteorological environmental parameters.

[0072] It should be noted that the fuzzy MACBETH method is a technique based on fuzzy mathematics and multi-attribute decision analysis. It helps identify and analyze the relative importance of different factors by transforming decision-makers' verbal preferences into numerical weights. Through semantic scales, decision-makers express the importance of various meteorological and environmental parameters, and these preferences are used for subsequent fuzzy inference and weight calculation.

[0073] A semantic scale is a rating tool used to quantify decision-makers' subjective judgments on the relative importance of various meteorological and environmental parameters. Typically, a semantic scale contains different descriptive labels to represent different weight levels. Common semantic scales are shown in Table 1:

[0074] Table 1 Semantic Scale

[0075]

[0076] Decision-makers assign a semantic scale value to the importance of each meteorological environmental parameter in different contexts, based on experience or expert knowledge. For example, in the measurement data of a photovoltaic power plant, solar irradiance (SI) may be considered the most important, while wind speed (W) may be considered less important in some cases.

[0077] S1212. For each decision-maker, construct a fuzzy comparison matrix of meteorological and environmental parameters based on the semantic scale.

[0078] Specifically, based on the ratings in the semantic scale, decision-makers compare the relative importance of various meteorological parameters. Typically, the comparison matrix is ​​a symmetric matrix where each pair of parameters has a rating indicating their relative importance. For example, if a decision-maker considers solar irradiance (SI) more important than ambient temperature (T), the corresponding item in the comparison matrix will have a larger value.

[0079] S1213. Construct a fuzzy linear programming model with the objective function of minimizing the sum of squares of the differences in the weights of all meteorological and environmental parameters.

[0080] It should be noted that there are n meteorological environmental parameters, such as r1, r2, ..., r m Let represent the fuzzy weights of each meteorological environmental parameter. Assume a decision-maker has assigned a score to the relative importance of these parameters (through fuzzification), and needs to optimize the weight of each meteorological parameter based on these scores. The objective function is to minimize the sum of squares of the differences in the weights of each meteorological environmental parameter, typically expressed as:

[0081]

[0082] After setting the objective function as minimizing the sum of squares of the differences in the weights of all meteorological and environmental parameters, additional constraints are required. These constraints typically include the following:

[0083] Normalization constraint: The sum of the weights of all meteorological and environmental parameters should be 1.

[0084] Nonnegativity constraint: Each weight should be nonnegative, that is:

[0085] A fuzzy linear programming model is constructed using an objective function and constraints to solve for the weights of meteorological and environmental parameters.

[0086] S1214. By solving the fuzzy linear programming model, the fuzzy weights of each meteorological environmental parameter are obtained, and the fuzzy weights are normalized to obtain the final weights of the meteorological environmental parameters.

[0087] Specifically, to solve this fuzzy linear programming model, standard linear programming methods can be used. The model is solved using constrained least squares (minimizing the difference of squares). The specific steps are as follows:

[0088] Constructing the constraint matrix: Transforming the constraints into matrix form facilitates subsequent linear programming solutions.

[0089] Solving the objective function: The objective function is a quadratic objective function, but since the constraints are linear, it can be transformed into a linear programming problem through some transformations.

[0090] Solving for weights: Use linear programming optimization algorithms (such as the simplex method or interior point method) to solve for the weight values.

[0091] S122. For each meteorological environmental parameter, calculate the niche overlap between measurement data under different conditions and measurement data under the same conditions, based on the weight of the meteorological environmental parameter.

[0092] In a preferred embodiment, the step of calculating the niche overlap between measurement data under different conditions and measurement data under the same conditions, based on the weight of each meteorological environmental parameter, includes:

[0093] S1221. Standardize the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions respectively.

[0094] It should be noted that the purpose of standardization is to eliminate dimensional differences between different meteorological environmental parameters, so that each parameter can be compared on the same scale. Standardization typically employs the Z-score standardization method.

[0095] S1222. For each meteorological environmental parameter, the measurement data under the same conditions shall be regarded as the ecological niche space.

[0096] It should be noted that in niche theory, niche space refers to a multidimensional space of a set of variables (in this case, meteorological and environmental parameters and measurement data). Here, measurement data under the same conditions are used as a standard "niche space," and then other measurement data are evaluated to determine how they overlap with this standard niche space. Specifically, this includes the following steps:

[0097] (1) Select measurement data under the same conditions:

[0098] The measurements were taken under relatively constant meteorological conditions (e.g., relatively stable parameters such as solar irradiance and temperature). These data will serve as the benchmark for the evaluation.

[0099] (2) Constructing ecological niche space:

[0100] Using standardized measurement data under the same conditions as coordinate axes, a multi-dimensional space is constructed. Each measurement parameter (such as current, voltage, etc.) can be regarded as a dimension, and the distribution of data points in this space reflects the characteristics of the measurement data under this meteorological condition.

[0101] S1223. For measurement data under different conditions, the degree of overlap between the measurement data under the same conditions and the measurement data under the same conditions in the niche space is calculated using cosine similarity to obtain the niche overlap degree.

[0102] Specifically, to calculate niche overlap, it is first necessary to calculate the similarity in niche space between measurement data under different conditions and measurement data under the same conditions. This is typically done using cosine similarity to quantify the degree of overlap.

[0103] Definition of cosine similarity:

[0104] Cosine similarity is used to calculate the similarity between two vectors (in this case, standardized vectors of measurement data), and the formula is:

[0105] Cosine Similarity=A·B / (||A||||B||);

[0106] Where: A and B represent two vectors (standardized measurement data under the same and different conditions, respectively); ||A|| and ||B|| represent the norms (i.e., the lengths of the vectors) of vectors A and B, respectively.

[0107] The cosine similarity value ranges from [-1, 1], where:

[0108] 1 indicates complete overlap, meaning the two vectors are exactly the same;

[0109] 0 indicates that the two vectors do not overlap at all, meaning they are orthogonal.

[0110] -1 indicates the exact opposite.

[0111] In the calculation of niche overlap, the absolute value of cosine similarity is usually taken because negative values ​​have no practical meaning in niche overlap.

[0112] S123. Using the Pearson correlation coefficient method, calculate the correlation coefficient between the measurement data and various meteorological environmental parameters to obtain the correlation analysis results.

[0113] Specifically, the Pearson correlation coefficient is a statistical method used to measure the strength and direction of the linear relationship between two variables. It quantifies the correlation between meteorological environmental parameters (such as temperature and humidity) and measured data (such as current and voltage). The meaning of the Pearson correlation coefficient:

[0114] q=1: Perfect positive correlation, the two variables are linearly positively correlated, as one variable increases, the other variable also increases.

[0115] q = -1: Perfectly negative correlation, the two variables are linearly negatively correlated, as one variable increases, the other variable decreases.

[0116] q = 0: There is no linear correlation; there is no obvious linear relationship between the two variables.

[0117] S124. Based on the results of niche overlap and correlation analysis, comprehensively evaluate the impact of each meteorological parameter on the uncertainty of measurement data, and determine the key meteorological parameters.

[0118] Specifically, by combining the results of niche overlap and correlation analysis, the impact of various meteorological parameters on the uncertainty of measurement data can be comprehensively evaluated from the following aspects:

[0119] Overlap and correlation combined: If the measurement data corresponding to a certain meteorological parameter have high overlap in the niche space and a large Pearson correlation coefficient (whether positive or negative), then the meteorological parameter has a small impact on the measurement data, and the measurement uncertainty is relatively low. If the measurement data corresponding to a certain meteorological parameter have low overlap in the niche space and a large Pearson correlation coefficient, then the meteorological parameter has a large impact on the measurement data uncertainty. If the Pearson correlation coefficient of a certain meteorological parameter is low (close to 0), regardless of its overlap, the meteorological parameter's impact on the measurement data uncertainty may be small.

[0120] Based on a combination of niche overlap and Pearson correlation coefficient, key meteorological parameters are those meteorological environmental parameters that have a significant impact on the uncertainty of measurement data and can substantially affect the measurement results. These meteorological parameters typically have the following characteristics:

[0121] Low niche overlap: This indicates that the relationship between changes in the meteorological parameter and the measurement data is more sensitive, which can easily lead to increased uncertainty in the measurement data.

[0122] A high Pearson correlation coefficient indicates a close relationship between the meteorological parameter and the measurement data; therefore, its changes may directly affect the stability and accuracy of the measurement data.

[0123] By identifying meteorological parameters that exhibit low overlap in niche overlap analysis but high correlation in correlation analysis, these parameters can be considered key meteorological parameters affecting the uncertainty of measurement data.

[0124] S13. Select measurement data under different key meteorological parameters, analyze the probability distribution law of the measurement data under key meteorological parameters through data fitting technology, and establish a comprehensive comparison relationship and mapping function of the measurement data based on the probability distribution law.

[0125] Specifically, based on the previous analysis, the key meteorological parameters affecting the uncertainty of the measurement data have been identified. Measurement data under these key meteorological parameters are selected as the analysis objects.

[0126] The analysis of probability distribution patterns using data fitting techniques includes the following steps:

[0127] Choose an appropriate probability distribution model based on the characteristics of the measurement data and the analysis requirements; common probability distribution models include normal distribution, log-normal distribution, exponential distribution, Weibull distribution, etc.

[0128] Use statistical software (such as Python's SciPy library, R language, etc.) to fit the measurement data and determine the parameters of the probability distribution model.

[0129] To evaluate the fit, methods such as the chi-square test and the KS test are commonly used to verify the goodness of fit.

[0130] Based on the fitting results, analyze the probability distribution pattern of the measurement data, understand the statistical characteristics of the data such as mean, variance, skewness, and kurtosis, as well as the distribution pattern of the data (such as symmetrical distribution, skewed distribution, etc.).

[0131] Furthermore, the comprehensive comparison relationship describes the relationship between measurement data under the same conditions and under different conditions; the comprehensive comparison relationship can be defined by comparing the probability distribution, statistical characteristics (such as mean, variance, etc.) or specific indicators (such as niche overlap) of two datasets.

[0132] Mapping functions are used to transform measurement data under different conditions into a form comparable to measurement data under the same conditions. Based on the comprehensive comparison relationship, linear or nonlinear mapping functions can be established. For example, if two datasets have similar probability distributions but different means, a linear mapping function can be established to adjust the means; if the distribution shapes are different, a more complex nonlinear mapping function may be needed.

[0133] In practical applications, mapping functions are used to transform measurement data under different conditions into a form comparable to measurement data under the same conditions; this helps to compare and analyze data under a unified standard, improving the comparability and usability of the data.

[0134] S2. Based on the electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the standard uncertainty under different evaluation results, and calculate the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty;

[0135] In a preferred embodiment, the step of calculating the standard uncertainty under different evaluation results based on electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, and calculating the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty includes:

[0136] S21. Based on the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions, calculate the standard uncertainty of the measurement data under the same meteorological conditions and the standard uncertainty of the measurement data under different single meteorological conditions respectively.

[0137] It should be noted that the calculation of measurement uncertainty can be divided into Type A (under the same meteorological conditions) and Type B (under different single meteorological conditions). Type A measurement results are evaluated using the experimental standard deviation, i.e., by independently repeating the experiment on the same measured data. Type B evaluation is based on the standard deviation of the probability distribution according to experience or assumptions, analyzing the probability distribution of the measurement results under actual influencing parameters, and determining the confidence interval (a, -a) of the possible values ​​of the measured electrical quantity.

[0138] In a preferred embodiment, the formulas for calculating the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions are as follows:

[0139]

[0140] u B (x) = a / k;

[0141]

[0142] In the formula, u A (x) represents the standard uncertainty of measurement data under the same meteorological conditions, X represents the average value of measurement data under the same meteorological conditions, n-1 represents the degrees of freedom, and n represents the number of measurement data under the same meteorological conditions. i u represents the i-th measurement data in the measurement data under the same meteorological conditions. B (x) represents the standard uncertainty of measurement data under different conditions and a single meteorological environment, a represents the half width of the possible value interval of the measurand, and k represents the coverage factor.

[0143] S22. Based on the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions, and combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the combined standard uncertainty.

[0144] In a preferred embodiment, the formula for calculating the combined standard uncertainty is:

[0145]

[0146] In the formula, u c (y) represents the combined standard uncertainty, u(x) i ) represents the input quantity x i The standard uncertainty, f, represents the comprehensive comparison relationship and mapping function of the measurement data, u(x) i ,x j ) represents the input quantity x i With input quantity x j The covariance, where N represents the total number of measurement data inputs, x j Represents the relationship with input quantity x i The j-th unrelated measurement data, where x represents the measurement data.

[0147] Furthermore, since there is no correlation between the same electrical quantities, therefore:

[0148]

[0149] In the formula, u c (y) represents the combined standard uncertainty, u i Let u represent the standard uncertainty component of the i-th input quantity. A u represents the standard uncertainty of measurement data under the same meteorological conditions. B N represents the standard uncertainty of measurement data under different single meteorological conditions, and i represents the ordinal number of measurement data inputs.

[0150] S23. Based on the combined standard uncertainty, calculate the effective degrees of freedom of the combined standard uncertainty using the formula for calculating effective degrees of freedom.

[0151] In a preferred embodiment, the formula for calculating the effective degrees of freedom is:

[0152]

[0153] In the formula, γ eff The effective degrees of freedom, u, represent the combined standard uncertainty. c (y) represents the combined standard uncertainty, u i (y) represents the standard uncertainty component of the i-th input quantity, γ i Let represent the degrees of freedom of the i-th measurement data input, N represent the total number of measurement data inputs, and i represent the ordinal number of the measurement data input.

[0154] S3. Utilize the effective degrees of freedom of the combined standard uncertainty, and calculate the environmental correction coefficient of the measurement data by looking up the table in conjunction with the confidence level. Apply the calculated environmental correction coefficient to the measurement data to adjust the measurement data.

[0155] In a preferred embodiment, the step of utilizing the effective degrees of freedom of the combined standard uncertainty, combining confidence level lookup table calculation to determine the environmental correction coefficient for the measurement data, and applying the calculated environmental correction coefficient to the measurement data to adjust the measurement data includes:

[0156] S31. Determine the expansion factor by looking up the table based on the effective degrees of freedom and confidence level of the combined standard uncertainty;

[0157] It should be noted that the expansion factor is obtained by looking up the table of standard uncertainty expansion factors based on the effective degrees of freedom and confidence level of the combined standard uncertainty. This expansion factor is used to convert the combined standard uncertainty into expanded uncertainty, so that the measurement uncertainty can be more accurately reflected at a specific confidence level.

[0158] First, you need to calculate or provide the effective degrees of freedom for the measurement data. Effective degrees of freedom are typically calculated from multiple factors, particularly the combined effect of the standard uncertainty of the input quantities and the degrees of freedom.

[0159] Choose a confidence level: Select an appropriate confidence level based on the application scenario. Common confidence levels include:

[0160] 68.27% (corresponding to 1 standard deviation, typically used in applications with lower confidence levels);

[0161] 95.45% (corresponding to 2 standard deviations, widely used in most applications);

[0162] 99.73% (corresponding to 3 standard deviations, applicable to situations with extremely high confidence);

[0163] Find the expansion factor from the table: Based on the known effective degrees of freedom and confidence level, find the corresponding value from the standard uncertainty expansion factor table.

[0164] For example, the value of the expansion factor depends on the effective degrees of freedom and the chosen confidence level. Generally, the value of the expansion factor decreases as the effective degrees of freedom increase and increases as the confidence level increases. Expansion factor tables are typically used to find appropriate values ​​for different effective degrees of freedom and confidence levels. These tables usually list some common effective degrees of freedom values ​​and their corresponding expansion factors. For example, Table 2 shows one such table.

[0165] Table 2 Confidence Level Lookup Table

[0166]

[0167] The table lists common effective degrees of freedom and their corresponding expansion factors at different confidence levels. Assuming an effective degree of freedom of 10 and a 95% confidence level, the expansion factor k′ = 7.0 can be found in the table.

[0168] S32. Calculate the expanded uncertainty of the measurement data based on the expansion factor and the combined standard uncertainty;

[0169] The formula for calculating the expanded uncertainty is as follows:

[0170] U=k′u c (y);

[0171] In the formula, U represents the expanded uncertainty, u c (y) represents the combined standard uncertainty, and k′ represents the expansion factor;

[0172] S33. Calculate the environmental correction factor for the testing instrument based on the expanded uncertainty;

[0173] It should be noted that the environmental correction factor is used to adjust the original measurement data, taking into account the influence of meteorological conditions on the measurement results. Its calculation formula is as follows:

[0174] Cenv=U / g avg ;

[0175] Where Cenv represents the environmental correction factor, indicating the adjustment factor required due to changes in meteorological environment; U represents the expanded uncertainty, which is calculated based on the combined standard uncertainty, representing the total uncertainty of the measurement data at a given confidence level; and g represents the total uncertainty of the measurement data at a given confidence level. avg This represents the average value of the measured data, that is, the arithmetic mean of all measurement results.

[0176] S34. Apply the calculated environmental correction factor to the original measurement data to adjust the measurement data.

[0177] Specifically, after calculating the environmental correction factor, it can be applied to adjust the original measurement data g to obtain the adjusted measurement data g′.

[0178] The formula is adjusted as follows:

[0179] g′=g·(1+Cenv);

[0180] Where: g′ represents the adjusted measurement data, that is, the measurement result after meteorological environment correction; g represents the original measurement data.

[0181] In summary, by utilizing the above-mentioned technical solutions of this invention, this invention, based on thorough theoretical analysis and mutual verification with standard laboratory results, refines and improves models and parameters using advanced numerical processing and analysis techniques, establishes confidence assessment criteria, and proposes a confidence assessment method for the field test indicators of photovoltaic power plants and theoretical analysis errors based on comparative statistical analysis. This method ensures the dynamic compatibility of uncertainty and data accuracy in field measurement results; it also effectively assesses field measurement errors, ensuring the reliability of photovoltaic power plant output parameters; and by forming environmental correction coefficients for measuring instruments, it proposes an evaluation method for the measurement uncertainty introduced by field measuring instruments in photovoltaic power plants.

[0182] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0183] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the confidence level of errors between field test indicators and theoretical analysis, characterized in that, include: S1. Based on meteorological environmental data and electrical quantity measurement data of photovoltaic power plants, measurement data under the same meteorological conditions and under different single meteorological conditions are selected. The final weights of meteorological environmental parameters are determined using the fuzzy MACBETH method based on these data. For each meteorological environmental parameter, the niche overlap between measurement data under different conditions and measurement data under the same conditions is calculated according to the parameter's weight. The correlation coefficient between the measurement data and each meteorological environmental parameter is calculated using the Pearson correlation coefficient method to obtain the correlation analysis results. Combining the niche overlap and correlation analysis results, the impact of each meteorological parameter on the uncertainty of the measurement data is comprehensively evaluated, and key meteorological parameters are determined. Measurement data under different key meteorological parameters are selected, and the probability distribution law of the measurement data under the key meteorological parameters is analyzed using data fitting techniques. Based on the probability distribution law, a comprehensive comparison relationship and mapping function for the measurement data are established. S2. Based on the electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the standard uncertainty under different evaluation results, and calculate the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty; S3. Utilize the effective degrees of freedom of the combined standard uncertainty, and calculate the environmental correction coefficient of the measurement data by looking up the table in conjunction with the confidence level. Apply the calculated environmental correction coefficient to the measurement data to adjust the measurement data.

2. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 1, characterized in that, The determination of the final weights of meteorological environmental parameters using the fuzzy MACBETH method, based on measurement data under identical meteorological conditions and measurement data under different single meteorological conditions, includes: S1211. Based on measurement data under the same meteorological conditions and measurement data under different single meteorological conditions, the semantic scale of the fuzzy MACBETH method represents the decision-maker's subjective judgment on the importance of different meteorological environmental parameters. S1212. For each decision-maker, construct a fuzzy comparison matrix of meteorological and environmental parameters based on the semantic scale. S1213. Construct a fuzzy linear programming model with the objective function of minimizing the sum of squares of the differences in the weights of all meteorological and environmental parameters. S1214. By solving the fuzzy linear programming model, the fuzzy weights of each meteorological environmental parameter are obtained, and the fuzzy weights are normalized to obtain the final weights of the meteorological environmental parameters.

3. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 2, characterized in that, For each meteorological environmental parameter, the calculation of the niche overlap between measurement data under different conditions and measurement data under the same conditions, based on the weight of the meteorological environmental parameter, includes: S1221. Standardize the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions respectively. S1222. For each meteorological environmental parameter, the measurement data under the same conditions shall be regarded as the ecological niche space. S1223. For measurement data under different conditions, the degree of overlap between the measurement data under the same conditions and the measurement data under the same conditions in the niche space is calculated using cosine similarity to obtain the niche overlap degree.

4. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 1, characterized in that, The step of calculating the standard uncertainty under different evaluation results based on electrical quantity measurement data, combined with the comprehensive comparison relationship and mapping function of the measurement data, and calculating the effective degrees of freedom of the combined standard uncertainty based on the standard uncertainty includes: S21. Based on the measurement data under the same meteorological conditions and the measurement data under different single meteorological conditions, calculate the standard uncertainty of the measurement data under the same meteorological conditions and the standard uncertainty of the measurement data under different single meteorological conditions respectively. S22. Based on the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions, and combined with the comprehensive comparison relationship and mapping function of the measurement data, calculate the combined standard uncertainty. S23. Based on the combined standard uncertainty, calculate the effective degrees of freedom of the combined standard uncertainty using the formula for calculating effective degrees of freedom.

5. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 4, characterized in that, The formulas for calculating the standard uncertainty of measurement data under the same meteorological conditions and the standard uncertainty of measurement data under different single meteorological conditions are as follows: ； ； In the formula, u A (x) represents the standard uncertainty of measurement data under the same meteorological conditions. Let x represent the average value of measurement data under the same meteorological conditions, n-1 represent the degrees of freedom, and n represent the number of measurement data under the same meteorological conditions. i u represents the i-th measurement data in the measurement data under the same meteorological conditions. B (x) represents the standard uncertainty of measurement data under different conditions and a single meteorological environment, a represents the half width of the possible value interval of the measurand, and k represents the coverage factor.

6. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 4, characterized in that, The formula for calculating the combined standard uncertainty is as follows: ； In the formula, u c (y) represents the combined standard uncertainty, u(x) i ) represents the input quantity x i The standard uncertainty, f, represents the comprehensive comparison relationship and mapping function of the measurement data, u(x) i ,x j ) represents the input quantity x i With input quantity x j The covariance, where N represents the total number of measurement data inputs, x j Represents the relationship with input quantity x i The j-th unrelated measurement data, where x represents the measurement data.

7. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 1, characterized in that, The formula for calculating the effective degrees of freedom is: ； In the formula, γ eff The effective degrees of freedom, u, represent the combined standard uncertainty. c (y) represents the combined standard uncertainty, u i (y) represents the standard uncertainty component of the i-th input quantity, γ i Let represent the degrees of freedom of the i-th measurement data input, N represent the total number of measurement data inputs, and i represent the ordinal number of the measurement data input.

8. The method for evaluating the confidence level of field test indicators and theoretical analysis errors according to claim 1, characterized in that, The process of utilizing the effective degrees of freedom of the combined standard uncertainty, combined with confidence level lookup table calculation of the environmental correction coefficient for the measurement data, and applying the calculated environmental correction coefficient to the measurement data to adjust the measurement data includes: S31. Determine the expansion factor by referring to a table based on the effective degrees of freedom and confidence level of the combined standard uncertainty; S32. Calculate the expanded uncertainty of the measurement data based on the expansion factor and the combined standard uncertainty; The formula for calculating the expanded uncertainty is as follows: U=k′u c (y); In the formula, U represents the expanded uncertainty, u c (y) represents the combined standard uncertainty, and k′ represents the expansion factor; S33. Calculate the environmental correction factor for the testing instrument based on the expanded uncertainty; S34. Apply the calculated environmental correction factor to the original measurement data to adjust the measurement data.

Images