Wind turbine generator combined simulation step length evaluation method

Abstract

The invention discloses a wind turbine generator combined simulation step length evaluation method. According to the method, simulation precision evaluation indexes are further divided into two aspects of overall similarity and transient difference, calculation is carried out through residual similarity and a feature selection verification method, and calculation results are converted into the same range and are weighted and merged. When simulation precision evaluation is carried out on a plurality of different parameters of joint simulation under the same step length, simulation precision evaluation indexes of the different parameters are combined in a weighting mode. According to the method, the difference of the joint simulation speed under different simulation step lengths is considered, the simulation speed is incorporated into an evaluation system, and the simulation speed and the simulation precision evaluation index are weighted and combined into a comprehensive evaluation index considering the simulation precision and the speed, so that joint simulation results with different step lengths are evaluated more comprehensively.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power joint simulation of new energy, and in particular relates to a step-length evaluation method for joint simulation of wind turbines. Background Art

[0002] With the continuous development of new energy wind power, its penetration rate in the power grid continues to increase, and the capacity of a single unit is gradually increasing. As a result, the load on wind turbines continues to increase, which is even more serious under various disturbances, posing a threat to the safe and stable operation of wind turbines.

[0003] Many studies on wind turbines focus on their electrical characteristics, so the modeling of the mechanical parts is simplified.

[0004] As research continues to deepen, the demand for models that take into account both the electrical characteristics and mechanical parts of wind turbines is also growing, leading to the application and development of wind turbine joint simulation technology.

[0005] In the research of wind turbines, modeling software such as Matlab / Simulink and PSCAD are usually used. However, these software focus more on modeling and simulation at the electrical level, and are insufficient in modeling the mechanical part. Software such as GH Bladed and FAST have powerful mechanical load analysis functions, but the modeling at the electrical level is relatively simple. Therefore, in order to combine the advantages of the software, joint simulation technology is used to establish a more comprehensive simulation model that takes into account both electrical and mechanical characteristics.

[0006] Residual similarity is a method used to analyze simulation accuracy, which reflects the similarity between the actual signal and the reference signal as a whole.

[0007] Feature selection validation analysis provides multi-angle evaluation results and can quantitatively evaluate the differences between data. Summary of the invention

[0008] The purpose of the present invention is to provide a wind turbine joint simulation step length evaluation method to comprehensively evaluate the selection of the joint simulation step length.

[0009] The residual similarity and feature selection verification analysis method in the present invention is widely used in the evaluation of signal accuracy.

[0010] The present invention further divides the evaluation of simulation accuracy into two parts: overall similarity and transient difference.

[0011] The present invention adopts an analysis method based on residual similarity to evaluate the overall similarity of joint simulation results under different simulation step sizes.

[0012] The present invention adopts an analysis method based on feature selection verification to evaluate the transient differences of joint simulation results under different simulation step sizes.

[0013] In the present invention, the index values ​​calculated by the residual similarity analysis method and the feature selection verification analysis method are unified into the same range and weightedly combined as the evaluation index of simulation accuracy.

[0014] In the present invention, when the simulation accuracy is evaluated for a plurality of different parameters of the joint simulation at the same step length, the simulation accuracy evaluation indicators of the different parameters are combined in a weighted manner.

[0015] The present invention introduces simulation speed into the evaluation system. The simulation speed evaluation index is calculated by the ratio of the actual simulation time to the reference simulation time and is weighted and combined with the simulation accuracy evaluation index to form a comprehensive evaluation index that takes both simulation accuracy and speed into consideration.

[0016] The beneficial effects of the present invention are:

[0017] The present invention further subdivides the evaluation of simulation accuracy into two aspects: overall and transient, and fully combines the advantages of residual similarity and feature selection verification methods to make the accuracy evaluation of joint simulation results more comprehensive;

[0018] The present invention takes into account the differences in joint simulation accuracy and speed brought about by different simulation step sizes, and incorporates the evaluation of simulation speed into the evaluation system, thereby taking into account the evaluation of simulation accuracy and speed at the same time.

[0019] The present invention will be further described below in conjunction with specific implementations and accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a wind turbine joint simulation step-size evaluation method provided in this application; DETAILED DESCRIPTION

[0021] A framework of a wind turbine joint simulation step size evaluation method is as follows: Figure 1 As shown, the comprehensive evaluation index that takes into account both simulation accuracy and speed is composed of a simulation accuracy evaluation index and a simulation speed evaluation index.

[0022] The simulation accuracy evaluation index consists of the overall similarity and transient difference of the simulation results, which are calculated by the residual similarity analysis method and the feature selection verification analysis method respectively.

[0023] The evaluation index calculation formula of the residual similarity analysis method is:

[0024]

[0025] In the formula, y i , are the reference signal sequence and the actual simulation signal sequence respectively; x i is the similarity of the i-th sequence point; r i is the weight of the i-th sequence point; φ is the residual similarity.

[0026] The feature selection verification analysis method uses Fourier transform to separate the DC, low-frequency and high-frequency components of the actual signal and the reference signal, and calculates the amplitude difference, feature difference and global difference indicators through each frequency component.

[0027] Fast Fourier transform is used to separate the different frequency components of the signal, and the four lowest frequency points are taken as DC components. The remaining frequency points are taken to calculate the total signal energy, and 40% of the total signal energy is used to determine the dividing point between the low-frequency component and the high-frequency component.

[0028] After obtaining the dividing point of low-frequency and high-frequency components, the low-frequency and high-frequency components are filtered.

[0029] Perform inverse Fourier transform on the filtered frequency components to obtain DC components, low-frequency components and high-frequency components.

[0030] The calculation formula for the amplitude difference of a single point is:

[0031]

[0032] Where ADM(k) is the amplitude difference of a single point; L real , L ref are the low-frequency components of the actual signal and the reference signal respectively; DC real 、DC ref are the DC components of the actual signal and the reference signal respectively; N L 、N DC are the low frequency and DC component data lengths respectively.

[0033] The calculation formula for the feature difference of a single point is:

[0034]

[0035] Where FDM(k) is the feature difference of a single point; L′ real , L′ ref are the first-order derivatives of the low-frequency components of the actual signal and the reference signal respectively; H′ real , H″ real and H′ ref , H″ ref are the first-order and second-order derivatives of the high-frequency components of the actual signal and the reference signal, respectively; N L′ 、N H′ 、N H″They are the data lengths of the first-order derivative of the low-frequency component, the first-order derivative of the high-frequency component, and the second-order derivative.

[0036] The global difference of a single point is calculated by the amplitude difference and feature difference of the single point. The calculation formula is:

[0037]

[0038] After calculating the single-point index, the overall index of amplitude difference, characteristic difference and global difference can be calculated. The calculation formula is:

[0039]

[0040] Where N ADM 、N FDM 、N GDM They are the data lengths of amplitude difference, feature difference, and global difference respectively.

[0041] In the present invention, the smaller the calculation result, the smaller the deviation and the higher the accuracy. The calculation result of the residual similarity is converted, and the calculation formula is:

[0042]

[0043] Where G is the overall difference index calculated from a certain parameter; φ is the residual similarity of the parameter.

[0044] In the present invention, the transient process required to be analyzed by the signal is evaluated by the feature selection verification method, the calculated global difference is used as the evaluation index, and it is normalized to convert the numerical range to be consistent with the overall difference index. The conversion formula is:

[0045]

[0046] The calculation formula of the simulation accuracy evaluation index of the joint simulation result in the present invention is:

[0047] X=aG+bF

[0048] Where X is the simulation accuracy evaluation index for the joint simulation results; a and b are the weights of the overall similarity and transient difference, respectively, and a+b=1.

[0049] When evaluating multiple different parameters of joint simulation under the same step length, the calculation formula of the simulation accuracy evaluation index is:

[0050] X=p1X1+p2X2+…+p n X n

[0051] In the formula, X1 to X nare simulation accuracy evaluation indicators of n different parameters; p1 to p n are the weights of different parameters, and their sum is 1.

[0052] In the present invention, the time used for the joint simulation of the reference step length is used as a benchmark, and the calculation formula for the evaluation index of the simulation speed is:

[0053]

[0054] Where, T is the simulation speed evaluation index; T real is the actual simulation time; T ref is the time taken for the reference simulation.

[0055] In the present invention, the calculation formula for the comprehensive evaluation index of multiple parameters of joint simulation under a certain step length, which takes into account both simulation accuracy and speed, is:

[0056] C=cX+dT

[0057] Where C is a comprehensive evaluation index that takes into account both simulation accuracy and speed; c and d are the weights of simulation accuracy and simulation speed, respectively, and the sum of c and d is 1.

[0058] In the present invention, the smaller the comprehensive evaluation index is, the higher the joint simulation accuracy of the corresponding simulation step is, the shorter the simulation time is, and the faster the speed is.

[0059] The above description is only for describing the specific implementation of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A wind turbine joint simulation step size evaluation method, characterized in that: At the same time, the joint simulation results are evaluated from two aspects: simulation accuracy and simulation speed. The simulation results are evaluated more comprehensively through comprehensive evaluation indicators that take into account both simulation accuracy and simulation speed.

2. A wind turbine joint simulation step length evaluation method according to claim 1, characterized in that: The evaluation index of simulation accuracy consists of the overall similarity and transient difference of simulation results, which are calculated by residual similarity analysis method and feature selection verification analysis method respectively.

3. A wind turbine joint simulation step length evaluation method according to claim 2, characterized in that: The index values ​​calculated by the residual similarity analysis method and the feature selection verification analysis method are unified into the same range and weightedly combined into the evaluation index of simulation accuracy.

4. A wind turbine joint simulation step length evaluation method according to claim 1, characterized in that: When evaluating the simulation accuracy of multiple different parameters of joint simulation under the same step length, the simulation accuracy evaluation indicators of different parameters are combined in a weighted manner.

5. A wind turbine joint simulation step length evaluation method according to claim 1, characterized in that: The evaluation index of simulation speed is calculated by the ratio of the actual simulation time to the reference simulation time.

6. A wind turbine joint simulation step length evaluation method according to claim 1, characterized in that: The calculated simulation accuracy evaluation index and simulation speed evaluation index are weighted and combined into a comprehensive evaluation index that takes both simulation accuracy and speed into consideration.

Images