Single-wind-field wind power output simulation method based on data mining

Abstract

The invention discloses a single-wind-field wind power output simulation method based on data mining. The method comprises the following steps: acquiring original data; filling missing values and repairing abnormal values; establishing a processing model for converting wind power output into wind speed; acquiring information such as cut-in wind speed, cut-out wind speed and rated power by using a data mining method; establishing a preliminary model based on Weibull distribution characteristics of wind speed and a random difference equation, and calculating to obtain K and C values of Weibull distribution and a parameter theta value in the random difference equation; based on the K, C and theta values in the step 5, generating future wind speed data in a time sequence by using a random difference equation model; and converting the generated wind speed data into wind power output data. The method has the beneficial effects that the output characteristics of the wind field and the fan estimation parameters are obtained only according to the historical data of the single wind field through a data mining method, so that future wind field output simulation is carried out, the time sequence and the distribution characteristics of the wind speed are considered at the same time, a user does not need to configure excessive extra parameters, and the method is suitable for practical engineering application.

Description

technical field [0001] The invention relates to a method for simulating wind power output, in particular to a method for simulating wind power output of a single wind farm based on data mining, and belongs to the technical field of power system planning. Background technique [0002] At present, the simulation of wind farm output is mainly divided into the following categories: [0003] 1. Ultra-short-term wind power forecasting based on wind power forecasting methods and forecasting models such as gray models can only do ultra-short-term forecasting, and cannot take into account the distribution and timing characteristics of wind speed; [0004] 2. Using random sampling in the Weibull distribution to obtain wind speed data can satisfy the distribution characteristics of wind speed, but lacks timing considerations, and the characteristic parameters of the Weibull distribution of the wind field must be known in advance; [0005] 3. Based on the time series ARIMA model method...

AI Technical Summary

Problems solved by technology

[0003] 1. Ultra-short-term wind power forecasting based on wind power forecasting methods, such as gray models, can only be done in the ultra-short term, and cannot take into account the distribution and timing characteristics of wind speed;

[0004] 2. Using random sampling in the Weibull distribution to obtain wind speed data can satisfy the distribution characteristics of wind speed, but lacks timing considerations, and the characteristic parameters of the Weibull distribution of the wind field must be known in advance;

[0005] 3. Based on the time series ARIMA model method, the temporal characteristics of wind speed are considered, but the distribution characteristics of wind speed are not considered

[0006] Among the above methods, which one cannot be used, it is necessary to strictly follow the model requirements and input a large number of model parameters in order to obtain a better simulation effect, which is far from the actual engineering application.

Images