Short-term wind power prediction method for optimizing SVM based on segmented ant colony algorithm

Abstract

The invention discloses a short-term wind power prediction method for optimizing an SVM based on a segmented ant colony algorithm. The method comprises the following steps: selecting and preprocessing original sample data to eliminate the influence of dimensions on the original data, and the original data comprising M sets of environmental influence factors influencing the wind power output power and the wind power output power corresponding to the environmental influence factors; determining the structure of the support vector machine according to the processed data; carrying out segmented ant colony optimization on the kernel function parameters and the penalty coefficient C of the support vector machine by combining the structure of the support vector machine to obtain an optimal parameter combination of the SVM meeting the expected error; and according to the obtained optimal parameter combination of the SVM, constructing an SVM model, that is, a wind power prediction model, and utilizing the model to realize short-term wind power prediction. According to the method, the influence of artificial subjective consciousness selection on model parameter combination can be effectively reduced, and the method is not limited to be applied to wind power generation power prediction and can be applied to model prediction of other known historical data.

Description

technical field [0001] The invention belongs to the technical field of wind power generation and grid connection, in particular to a short-term wind power prediction method based on segmented ant colony algorithm optimization SVM. Background technique [0002] Although the large-scale development of wind power has effectively alleviated the energy crisis and environmental pollution problems, due to many factors affecting wind energy, the output of wind turbines is random, fluctuating and unstable, which brings the characteristics of incomplete controllability. Large-scale wind power access has an impact on the stable operation and dispatch of the power system, so accurate short-term wind power forecasting is very important for improving the economic and stable operation of the power system. [0003] Support vector machines are widely used because of their excellent high-dimensional mapping capabilities. Among them, the kernel function of the support vector machine converts ...

AI Technical Summary

Problems solved by technology

[0005] The object of the present invention is to provide a segmented ant colony optimization support vector machine prediction method that solves the short-term wind power prediction cross-validation and the neural network is easily affected by subjective factors in parameter selection and falls into local optimum.

Images