Wind speed distribution fitting method, system and device based on adaptive golden sine differential evolution Weibull model and medium

Abstract

The invention discloses a wind speed distribution fitting method, system, equipment and medium based on a self-adaptive golden sine differential evolution Weibull model, and belongs to the technical field of wind speed distribution fitting. The wind speed distribution fitting method comprises the steps that wind speed observation data of a wind power plant are acquired, and the wind speed data are preprocessed and summarized; dividing the wind speed data into a frequency distribution format with a fixed wind speed as an interval; based on a wind speed frequency distribution format, establishing a wind speed probability distribution model which comprises a Weibull distribution model, a mixed two-component Weibull model and a model parameter value boundary constraint condition; optimizing the wind speed probability distribution model by adopting an evolutionary optimization algorithm fused with a self-adaptive disturbance mechanism; and evaluating an optimization result by using a plurality of goodness-of-fit test indexes, and selecting an optimal model parameter combination as a final wind speed modeling scheme according to an evaluation result. The method is simple and convenient in calculation process, higher in precision, low in cost and higher in engineering practical value.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind speed distribution fitting, and particularly relates to a wind speed distribution fitting method, system, device and medium based on an adaptive golden sine difference evolution Weibull model. BACKGROUND

[0002] Under the background of global energy transformation, wind energy, as a clean and renewable energy, is rapidly rising and becoming an important force to promote sustainable development. In recent years, with technological progress and policy support, the wind power industry has ushered in an unprecedented development opportunity. Wind energy is a clean and renewable energy with the advantages of less environmental pollution and lower development cost, and has become one of the leading clean energies. Wind energy will meet 25% to 33% of global electricity demand and will become one of the fastest growing renewable energies. In addition, compared with traditional fossil energy, wind energy shows great potential advantages and will play an important role in future energy development, and is a high-quality energy choice to achieve sustainable development.

[0003] However, the intermittency and randomness of wind energy also bring challenges to wind power development and utilization. In the design and planning process of wind energy projects, accurate assessment of wind energy resources is the basis for wind power projects. The wind resource level of a wind farm directly affects the early economic value, risk assessment, wind turbine selection and power generation estimation of the wind farm. Therefore, accurately describing the wind speed fluctuation characteristics is particularly critical, which has important guiding significance for the development and utilization of wind energy.

[0004] Currently, the research on wind speed probability distribution model is continuously deepened, and a series of mathematical statistical models for fitting wind speed are proposed, among which the Weibull distribution is the most widely used in wind resource assessment. However, with the rapid development of wind energy, a large number of models have been applied to wind speed modeling, which brings difficulties to the solution of parameters. The parameter extraction method for wind speed probability distribution model mainly includes numerical method and meta-heuristic optimization algorithm. The traditional numerical method is affected by the distribution format of wind speed data and the complexity of the model, and has poor adaptability and low parameter estimation accuracy. Compared with the numerical method, the meta-heuristic algorithm is not limited by conditions and can adapt to the parameter solution of various models with high calculation efficiency, but it is easy to fall into local optimum. In view of the above problems, the present application considers using an improved meta-heuristic algorithm to optimize the parameters of the wind speed probability model. Finally, the example verifies that the method has higher accuracy and better stability, and comprehensively considers the implementation difficulty and actual demand of the method. SUMMARY

[0005] In view of the above problems, the present application is proposed.

[0006] Therefore, the present application solves the technical problem of how to solve the problem of poor adaptability and low parameter estimation accuracy of traditional numerical methods affected by wind speed data distribution format and model complexity.

[0007] To solve the above technical problems, the present application provides the following technical solutions: a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model, which includes the following steps: obtaining wind speed observation data of a wind farm, preprocessing and summarizing the wind speed data, and dividing the wind speed data into a frequency distribution format with fixed wind speed intervals; based on the wind speed frequency distribution format, establishing a wind speed probability distribution model, including a Weibull distribution model, a mixed two-component Weibull model, and model parameter value boundary constraint conditions; using an evolutionary optimization algorithm with a fusion adaptive disturbance mechanism to optimize the wind speed probability distribution model, improving the accuracy of the wind speed fitting model; using multiple goodness-of-fit test indicators to evaluate the optimization results, and selecting the optimal model parameter combination as the final wind speed modeling scheme according to the evaluation results.

[0008] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the establishment of the wind speed probability distribution model includes the construction of candidate modeling structures based on the single Weibull model and the mixed two-component Weibull model, respectively.

[0009] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the evolutionary optimization algorithm includes dynamic updating of control parameters to achieve adaptive adjustment of the mutation degree in each iteration.

[0010] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the selection of the optimal model parameter combination as the final wind speed modeling scheme according to the evaluation results includes comprehensive evaluation of the modeling results based on different types of error measurement functions.

[0011] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the establishment of the wind speed probability distribution model based on the wind speed frequency distribution format includes modeling of the wind speed distribution by the scale parameter and the shape parameter of the Weibull distribution model, the mixed two-component Weibull model including two scale parameters, two shape parameters, and a weight parameter, the construction of an objective function based on the difference between the wind speed frequency distribution and the model prediction probability, and the setting of value range constraints for various parameters of the model.

[0012] The preferred scheme has stronger modeling adaptability when dealing with different types of wind speed scenarios by introducing two different structure wind speed probability distribution models, constructing an objective function according to the difference between the frequency distribution and the predicted probability in the modeling process, and setting a value range constraint for each model parameter.

[0013] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the evolution optimization algorithm includes a golden sine strategy with an adaptive disturbance mechanism based on the differential evolution algorithm, wherein the disturbance intensity gradually decreases with the increase of the iteration number, realizing the transition from global search to local development.

[0014] The preferred scheme introduces a golden sine strategy with an adaptive disturbance mechanism into the traditional differential evolution algorithm, so that the disturbance intensity can gradually decrease with the increase of the iteration number, thereby enhancing the global search ability in the early optimization stage, gradually focusing on local search in the later iteration stage, effectively reducing the risk of falling into local optimum, and improving the convergence efficiency and robustness of the overall algorithm in the wind speed model parameter optimization task.

[0015] As a preferred scheme of the wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model, the evaluation of the optimization results by using multiple goodness-of-fit test indicators includes that when the goodness-of-fit of the optimization results is evaluated, multiple goodness-of-fit test indicators are used to quantitatively judge the fitting degree between the model prediction results and the wind speed frequency distribution, and based on the judgment result, it is determined whether the current parameter combination is the optimal solution required for modeling.

[0016] Another object of the present application is to provide a wind speed distribution fitting system based on an adaptive golden sine differential evolution Weibull model.

[0017] To solve the above technical problems, the application provides the following technical scheme: a wind speed distribution fitting system based on an adaptive golden sine difference evolution Weibull model, comprising a data acquisition and processing module, a model construction module, a model optimization module and an evaluation module; the data acquisition and processing module is used to acquire wind speed observation data of a wind farm, and the wind speed data are preprocessed and summarized, and the wind speed data are divided into a frequency distribution format with fixed wind speed as an interval; the model construction module is used to establish a wind speed probability distribution model based on the wind speed frequency distribution format, including a Weibull distribution model, a mixed two-component Weibull model and a model parameter value boundary constraint condition; the model optimization module is used to optimize the wind speed probability distribution model by using an evolution optimization algorithm with a fusion adaptive disturbance mechanism, so as to improve the precision of the wind speed fitting model; and the evaluation module is used to evaluate the optimization result by using multiple fitting goodness test indexes, and the optimal model parameter combination is selected as a final wind speed modeling scheme according to the evaluation result.

[0018] The application provides a computer device, comprising a memory and a processor, and the memory stores a computer program, characterized in that the processor implements the steps of the wind speed distribution fitting method based on the adaptive golden sine difference evolution Weibull model when executing the computer program.

[0019] The application provides a computer readable storage medium, which stores a computer program, characterized in that the computer program implements the steps of the wind speed distribution fitting method based on the adaptive golden sine difference evolution Weibull model when executed by a processor.

[0020] The application has the beneficial effects that the application takes the difference between the actual wind speed frequency distribution and the probability density function as an objective function, takes the value range of the unknown parameters of the Weibull distribution model as a constraint, and uses an improved adaptive golden sine difference evolution algorithm to more accurately estimate the parameters of the Weibull distribution model. Meanwhile, the method is simple in calculation process, high in precision, low in cost and high in engineering practical value. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1A flow chart of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model is provided for an embodiment of the present application.

[0023] Figure 2 A flow chart of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model is provided for an embodiment of the present application.

[0024] Figure 3 A function curve fitting graph of a coastal area wind power plant A of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model is provided for an embodiment of the present application.

[0025] Figure 4 A function curve fitting graph of a coastal area wind power plant B of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model is provided for an embodiment of the present application.

[0026] Figure 5 A system scheme module graph of a wind speed distribution fitting system based on an adaptive golden sine differential evolution Weibull model is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0028] Embodiment 1, refer to Figure 1 For the first embodiment of the present application, the embodiment provides a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model, comprising:

[0029] S1, obtaining wind speed observation data of a wind power plant, and pre-processing and summarizing the wind speed data, and dividing the wind speed data into a frequency distribution format with fixed wind speed as an interval.

[0030] S2, based on the wind speed frequency distribution format, establishing a wind speed probability distribution model, including a Weibull distribution model, a mixed two-component Weibull model and a model parameter value boundary constraint condition.

[0031] S3, using an evolutionary optimization algorithm with a fusion adaptive disturbance mechanism to optimize the wind speed probability distribution model, and improving the accuracy of the wind speed fitting model.

[0032] S4, evaluate the optimization results by using multiple goodness-of-fit test indicators, and select the optimal model parameter combination as the final wind speed modeling scheme according to the evaluation results.

[0033] It should be noted that the wind speed data of the wind farm has the characteristics of strong continuity, violent fluctuation, irregular peak change and the like, and the traditional wind speed distribution modeling method based on static parameter configuration is difficult to obtain good modeling effect in multiple scenarios. Especially in the presence of complex wind field forms (such as bimodal wind speed distribution), the model fitting error is large, and the parameter adjustment cost is high. At the same time, the wind speed observation data often contains measurement errors and mutation points, which increases the difficulty of model construction and optimization.

[0034] The present application enhances the generality of the model by constructing a frequency distribution format in the data preprocessing stage, so that the original wind speed sequence participates in modeling in a discrete manner; adopts a combination of parameter modeling and boundary constraint to ensure the convergence and stability of the modeling structure; introduces an evolutionary optimization method with adaptive disturbance capability to improve the robustness of model parameter optimization; at the same time, multiple goodness-of-fit evaluation methods are used to comprehensively judge the model results, realizing a high-precision, scalable wind speed distribution fitting process.

[0035] Embodiment 2, refer to Figure 1 and Figure 2 The present application is a second embodiment based on the above embodiment, which provides a wind speed distribution fitting method based on an adaptive golden sine difference evolution Weibull model.

[0036] In the present application, wind speed observation data of a wind farm is obtained in step S1, and the wind speed data is preprocessed and summarized, and the wind speed data is divided into a frequency distribution format with a fixed wind speed interval.

[0037] Specifically, wind speed data of a wind farm is collected, preprocessed and summarized, and then the wind speed data is divided into a frequency distribution format with an interval of 1 m / s; in wind speed modeling, an interval of 1 m / s can simplify data statistics, probability analysis and model establishment.

[0038] In an alternative embodiment, the preprocessing operation is locally smoothed by a sliding window strategy, and at the same time, a fluctuation recognition algorithm based on statistical distribution characteristics is used to automatically mark and remove data mutation segments, so as to build a wind speed data distribution with higher continuity.

[0039] In another alternative embodiment, the preprocessing stage does not perform physical layer cleaning on the original data, but uses a distribution reconstruction method for data resampling, maps the wind speed values in the historical period to the interval frequency, and then directly models, so as to reduce the dependence on edge measurement points or packet loss data.

[0040] In the embodiment of the present application, the wind speed probability distribution model is established based on the wind speed frequency distribution format in step S2. The candidate modeling structure is constructed based on a single Weibull model and a mixed two-component Weibull model respectively.

[0041] The Weibull distribution model models the wind speed distribution through a scale parameter and a shape parameter, the mixed two-component Weibull model includes two scale parameters, two shape parameters and a weight parameter, a target function is constructed according to the difference between the wind speed frequency distribution and the predicted probability of the model, and the value range constraint of each type of parameter of the model is set.

[0042] According to the wind speed data frequency distribution format provided in step S1, an accurate wind speed probability distribution model is established, including a Weibull distribution model, a mixed two-component Weibull model, and a model parameter value boundary constraint condition.

[0043] The Weibull distribution probability density function is as follows:

[0044]

[0045] wherein f w (v) is the Weibull function probability density function, c is the scale parameter, k is the scale parameter, and v is the wind speed.

[0046] The mixed two-component Weibull distribution probability density function formula is as follows:

[0047]

[0048] wherein f W-W (v) is the mixed Weibull function probability density function, w is the weight coefficient, c1 is the first scale parameter, c2 is the second scale parameter, k1 is the first shape parameter, and k2 is the second shape parameter.

[0049] The model parameter value boundary constraint condition includes the construction of the target function and the model parameter constraint.

[0050] The target function is expressed as:

[0051]

[0052] wherein minRMSE(x) is the minimum root mean square error, N is the number of wind speed interval intervals, f m is the actual wind speed frequency, f m (v, x) is the probability function predicted probability, x includes all parameters to be solved, and the specific content is as follows:

[0053] For the Weibull distribution model:

[0054]

[0055] x={c,k}

[0056] For the mixed Weibull distribution model:

[0057]

[0058] x={w,c1,k1,c2,k2}

[0059] The setting range of Weibull parameters, the appropriate range obtained according to experimental results, and the range constraint of model parameters:

[0060] Weibull distribution model:

[0061] 0≤c≤30

[0062] 0≤k≤30

[0063] Wherein, c and k are the scale parameter and shape parameter of Weibull distribution respectively.

[0064] Mixed Weibull distribution model:

[0065] 0≤w≤1,0≤c1≤30,0≤k1≤30

[0066] 0≤c2≤30,0≤k2≤30

[0067] Wherein, w is the weight coefficient, c1 is the first scale parameter, c2 is the second scale parameter, k1 is the first shape parameter, and k2 is the second shape parameter.

[0068] In the embodiment of the application, the evolutionary optimization algorithm with fusion adaptive perturbation mechanism is used in step S3 to optimize the wind speed probability distribution model, so as to improve the accuracy of the wind speed fitting model.

[0069] The control parameters are dynamically updated in each iteration to realize adaptive adjustment of the mutation degree. The golden sine strategy with fusion adaptive perturbation mechanism is introduced on the basis of the differential evolution algorithm, wherein the perturbation intensity gradually decreases with the increase of the iteration number, so as to realize the transition from global search to local development.

[0070] Specifically, as shown in Figure 2 1) Collect wind speed data of the wind farm, pretreat the wind speed data and summarize.

[0071] 2) According to the collected wind speed historical data, the wind speed data is divided into frequency distribution format with 1m / s interval, and the wind speed probability distribution model is established.

[0072] 3) Generate initial population. According to the set population size, each individual in the population represents a set of model parameter solutions. Initialize the population within the constraint range, and calculate the fitness value of the individual according to the objective function.

[0073] 4) New mutated individual obtained according to the golden sine mutation strategy.

[0074] Golden sine mutation strategy:

[0075] v i = x r1 +F·(x r2 -x r3 )

[0076]

[0077] Adaptive dynamic adjustment parameters:

[0078]

[0079] Where, v i is the current mutated individual; v i new is the newly generated mutated individual; x best is the optimal individual in the current population; F is the scaling factor, taking 0.5; x r1 , x r2 , x r3 are three different experimental individuals, and satisfy r1≠r2≠r3≠i; θ1 and θ2 are the angles of trigonometric functions, θ1∈(0,2π), θ2∈(0,π); is the golden section ratio, α is the disturbance intensity; iter is the current iteration number, Max_iter is the maximum iteration number; α0=0.05, is the golden section ratio constant, and α0 is the disturbance intensity constant. It gradually decreases with the increase of the iteration number, gradually deviating to local development; α is weakened by the disturbance during the iteration process, balancing the global and local search of the algorithm.

[0080] The experimental vector is generated by the crossover operation of the differential evolution algorithm, and the expression is:

[0081]

[0082] Where, is the value of the ith individual in the jth dimension after the end of the tth generation crossover; is the value of the newly generated ith mutated individual in the jth dimension; is the value of the original ith individual in the jth dimension; rand jis a number between 0 and 1; CR is a cross factor, and the greater the value, the faster the convergence, and the easier the early maturation phenomenon, and the value is 0.9.

[0083] The generated experimental individual is compared with the original individual fitness according to the greedy criterion, and the next generation individual is updated with small fitness:

[0084]

[0085] wherein, U t i is the tth generation of the generated experimental individual, x i is the original individual, and f(·) is the objective function.

[0086] 5) Calculate the fitness function and keep the optimal individual.

[0087] 6) According to the set termination condition, it is judged whether to end the loop, if yes, the result is output, otherwise go to step 4).

[0088] In the embodiment of the application, the optimization result is evaluated by using multiple goodness-of-fit test indicators in step S4, and the modeling result is comprehensively evaluated based on different types of error measurement functions.

[0089] When the optimization result is evaluated for the goodness-of-fit of the model, multiple goodness-of-fit test indicators are used to quantitatively judge the fitting degree between the model prediction result and the wind speed frequency distribution, and based on the judgment result, it is determined whether the current parameter combination is the optimal solution required for modeling.

[0090] Specifically, the four goodness-of-fit test indicators of the application include root mean square error RMSE, mean absolute error MAE, chi-square test X 2 and coefficient of determination R 2 , and their characteristics are described as follows:

[0091]

[0092] wherein, y i is the predicted wind speed frequency based on the distribution function, is the actual wind speed frequency, is the average probability value, and wherein the values of RMSE, MAE, X 2 are closer to 0, indicating better fitting effect; the value of R 2 ranges from 0 to 1, and the value is closer to 1, indicating higher fitting precision.

[0093] In an alternative embodiment, the evaluation indicators include error measures such as bias coefficient, mean square log error, and residual sum of squares, which are used to focus on the assessment of the matching degree of the predicted probability in the extreme wind speed interval, so as to adapt to the fitting task of the multi-peak or extreme skewed wind speed distribution scenario.

[0094] In another alternative embodiment, the goodness-of-fit test does not use traditional statistical quantities, but uses distribution overlap measurement methods such as Kullback-Leibler divergence (KL divergence) or Bhattacharyya distance to directly measure the overall morphological difference between the model prediction distribution and the original wind speed distribution, and the parameter combination corresponding to the minimum divergence result is taken as the final modeling output.

[0095] The embodiments of the present application effectively cover multiple aspects such as error amplitude, fitting consistency, model reliability, and fitting significance through multi-dimensional statistical test methods, so that the optimal parameter combination finally selected not only satisfies local error optimization, but also has overall modeling stability and interpretability. Compared with the scheme relying on a single indicator, the fitting ability of the model to the wind speed characteristics is more comprehensively reflected, and the adaptability and prediction reliability of the final modeling result in complex wind field environment are improved.

[0096] Embodiment 3, refer to Figure 3 and Figure 4 The third embodiment of the present application provides a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model, in order to verify the beneficial effects of the present application, scientific demonstration is carried out through experiments.

[0097] For the embodiments of the present application, wind speed data of two wind farms A and B in the coastal area of China are used. The data collection time of the two wind farms A and B is from January 1, 2021 to January 1, 2023, a total of 19720 data. And Weibull distribution and mixed Weibull distribution model are used for experiment. The related parameters of the differential evolution algorithm are set as follows: the population size is set to 30, and the iteration number is set to 100 generations. The parameters of the improved differential evolution algorithm are consistent with the original algorithm. The improved DE is introduced into the original differential evolution algorithm, Weibull-Weibull is a mixed two-component Weibull.

[0098] The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model of the present application is adopted, and the optimal parameter results of the two models are shown in Tables 1 and 2, and the curve fitting of the two distribution functions is shown in Figure 3 、 Figure 4The optimization results of the original differential evolution algorithm and the improved differential evolution algorithm of the present application are compared in Table 3 and Table 4.

[0099] Table 1: Model parameter estimation results of wind farm A

[0100]

[0101] Table 2: Model parameter estimation results of wind farm B

[0102]

[0103] Table 3: Comparison of goodness-of-fit results of the original differential evolution algorithm and the improved differential evolution algorithm of wind farm A

[0104] Model Method RMSE MAE X 2 ]]> [R 2 ]]> Weibull DE 0.00925 0.00672 0.00171 0.93387 Weibull Improved DE 0.00900 0.00632 0.00163 0.93787 Weibull-Weibull DE 0.00494 0.00343 0.00070 0.98117 Weibull-Weibull Improved DE 0.00382 0.00248 0.00044 0.98870

[0105] Table 4: Comparison of goodness-of-fit results of the original differential evolution algorithm and the improved differential evolution algorithm of wind farm B

[0106] Model Method RMSE MAE X 2 ]]> [R 2 ]] Weibull DE 0.00938 0.00829 0.00124 0.92620 Weibull Improved DE 0.00922 0.00817 0.00112 0.93020 Weibull-Weibull DE 0.00417 0.00349 0.00020 0.98538 Weibull-Weibull Improved DE 0.00274 0.00204 0.00012 0.99372

[0107] From the optimized results, it can be seen that, in general, for two-parameter Weibull and five-parameter mixed Weibull, the present application performs better than the original method in the two wind farms, and the precision of the obtained parameters is higher.

[0108] Figure 3 、 Figure 4To improve the curve fitting effect of the differential evolution algorithm for two models, it can be clearly seen from the figure that the improved differential evolution algorithm can also accurately identify the five-parameter mixed distribution model. By comparing the fitting results, in wind farm A, the four indicators of using the original differential evolution algorithm to calculate the Weibull distribution parameters are: RMSE = 0.00925, MAE = 0.00672, X2 = 0.00171, R2 = 0.93387, and the four indicators of the improved differential evolution algorithm are: RMSE = 0.00900, MAE = 0.00632, X2 = 0.00163, R2 = 0.93787. Compared with the original differential evolution algorithm, the RMSE of the improved differential evolution algorithm is reduced by 2.9%, the MAE is reduced by 6.0%, the X2 is reduced by 4.7%, and the R2 is increased by 0.43%. For the mixed Weibull model, the difference between the two is more obvious, the RMSE of the improved differential evolution algorithm is reduced by 22.7%, the MAE is reduced by 27.7%, the X2 is reduced by 37.1%, and the R2 is increased by 0.76%. In addition, when using the original differential algorithm to extract parameters of the mixed Weibull distribution with more parameters, the fitting failure problem is prone to occur, which shows instability. The experimental conclusion of wind farm B is the same as that of wind farm A. Using the improved differential evolution algorithm to estimate the parameters of the model, the four indicators are better than those of the original differential evolution algorithm, and the optimization effect is obvious. The specific results are shown in Table 2.

[0109] The experiments of the two wind farms show that, compared with the original differential evolution algorithm, the adaptive golden sine differential evolution algorithm of the application effectively improves the accuracy of parameter estimation, is simple and efficient, and has strong adaptability.

[0110] In summary, using the improved adaptive golden sine differential evolution algorithm to solve the Weibull wind speed probability distribution model can improve the accuracy of the model to the actual wind speed distribution, so as to obtain the optimal wind speed modeling, which has a certain guiding role for predicting wind resources and reducing the uncertainty of wind energy in actual engineering practice.

[0111] Embodiment 4 is the fourth embodiment of the application, which is different from the first two embodiments:

[0112] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0113] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered a list of executable instructions for implementing logic functions, and can be specifically embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or in conjunction with these instructions execution systems, apparatuses, or devices. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport programs for use by an instruction execution system, apparatus, or device, or in conjunction with these instruction execution systems, apparatuses, or devices.

[0114] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by editing, interpreting, or otherwise processing, if necessary, in other suitable ways to be electronically obtained, and then stored in the computer memory.

[0115] It should be understood that various parts of the present application can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and as in another embodiment, it can be implemented using a combination of any of the following technologies known in the art: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0116] Embodiment 5, refer to Figure 5 As a fifth embodiment of the present application, the embodiment provides a wind speed distribution fitting system based on adaptive golden sine difference evolution Weibull model, comprising a data acquisition and processing module, a model construction module, a model optimization module and an evaluation module.

[0117] The data acquisition and processing module is used to obtain wind speed observation data of a wind farm, and to pre-process and aggregate the wind speed data, and to divide the wind speed data into a frequency distribution format with fixed wind speed intervals.

[0118] The model construction module is used to establish a wind speed probability distribution model based on the wind speed frequency distribution format, including a Weibull distribution model, a mixed two-component Weibull model, and a model parameter value boundary constraint condition.

[0119] The model optimization module is used to optimize the wind speed probability distribution model using an evolutionary optimization algorithm with a fusion adaptive disturbance mechanism, to improve the accuracy of the wind speed fitting model.

[0120] The evaluation module is used to evaluate the optimization results using a plurality of goodness-of-fit test indicators, and to select the optimal model parameter combination as the final wind speed modeling scheme according to the evaluation results.

[0121] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model, characterized by: include, Obtain wind speed observation data from the wind farm, preprocess and summarize the wind speed data, and divide the wind speed data into a frequency distribution format with fixed wind speed intervals; Based on the wind speed frequency distribution format, a wind speed probability distribution model is established, including the Weibull distribution model, the mixed two-component Weibull model, and the boundary constraints of the model parameters; An evolutionary optimization algorithm integrated with an adaptive perturbation mechanism is used to optimize the wind speed probability distribution model, thereby improving the accuracy of the wind speed fitting model. The optimization results were evaluated using multiple goodness-of-fit test indices, and the optimal model parameter combination was selected as the final wind speed modeling scheme based on the evaluation results.

2. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 1, characterized in that: The establishment of the wind speed probability distribution model includes constructing candidate modeling structures based on a single Weibull model and a mixed two-component Weibull model respectively.

3. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 2, characterized in that: The evolutionary optimization algorithm includes dynamically updating control parameters in each round of iteration to achieve adaptive adjustment of the degree of variation.

4. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 3, characterized in that: The selecting of the optimal model parameter combination as the final wind speed modeling scheme according to the evaluation results includes comprehensively evaluating the modeling results based on different types of error metric functions.

5. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 4, characterized in that: The wind speed probability distribution model is established based on the wind speed frequency distribution format, including: The Weibull distribution model models the wind speed distribution through scale parameters and shape parameters. The mixed two-component Weibull model includes two scale parameters, two shape parameters and a weight parameter. The objective function is constructed based on the difference between the wind speed frequency distribution and the model prediction probability, and the value range constraints are set for various parameters of the model.

6. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 4, characterized in that: The evolutionary optimization algorithm includes a golden sine strategy that introduces an adaptive perturbation mechanism based on a differential evolution algorithm, wherein the perturbation intensity gradually decreases with the increase in the number of iterations, thereby achieving a transition from global search to local development.

7. The wind speed distribution fitting method based on the adaptive golden sine differential evolution Weibull model according to claim 4, characterized in that: The optimization results are evaluated using multiple goodness of fit test indicators, including: When evaluating the model fitting goodness of the optimization results, multiple goodness-of-fit test indicators are used to quantitatively judge the degree of fit between the model prediction results and the wind speed frequency distribution, and based on the judgment results, it is determined whether the current parameter combination is the optimal solution required for modeling.

8. A wind speed distribution fitting system based on an adaptive golden sine differential evolution Weibull model, applying a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model as claimed in any one of claims 1 to 7, characterized in that: include: Data acquisition and processing module, model building module, model optimization module and evaluation module; The data acquisition and processing module is used to obtain wind speed observation data of the wind farm, pre-process and summarize the wind speed data, and divide the wind speed data into a frequency distribution format with fixed wind speed intervals; The model building module is used to establish a wind speed probability distribution model based on the wind speed frequency distribution format, including a Weibull distribution model, a mixed two-component Weibull model, and model parameter value boundary constraints; The model optimization module is used to optimize the wind speed probability distribution model using an evolutionary optimization algorithm integrated with an adaptive perturbation mechanism, thereby improving the accuracy of the wind speed fitting model; The evaluation module is used to evaluate the optimization results using multiple goodness of fit test indicators, and select the optimal model parameter combination as the final wind speed modeling solution based on the evaluation results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a wind speed distribution fitting method based on an adaptive golden sine differential evolution Weibull model according to any one of claims 1 to 7 are implemented.

Images