Short-term wind speed prediction method based on variational heteroscedasticity Gaussian process regression

Abstract

The invention relates to the technical field of short-term wind speed prediction, and discloses a short-term wind speed prediction method based on variational heteroscedasticity Gaussian process regression. The method comprises the following steps: firstly, decomposing an original wind speed time sequence into a sub-sequence set with stronger regularity by adopting complete set empirical mode decomposition with adaptive noise, calculating a partial autocorrelation function value of each sub-sequence, and selecting a significant lag time sequence at a confidence level of 95% as an input variable; secondly, training and predicting each subsequence by adopting variational heteroscedasticity Gaussian process regression, and finally, combining prediction results of all the subsequences to obtain the final prediction result of the wind speed time sequence. Compared with the prior art, the wind speed time sequence is predicted by adopting the variational heteroscedasticity Gaussian process regression model, the prediction capability is stronger, the performance of the variational heteroscedasticity Gaussian process regression model is superior to that of a standard Gaussian process regression model, and higher prediction precision can be obtained.

Description

technical field [0001] The invention relates to the field of short-term wind speed prediction, in particular to a short-term wind speed prediction method based on variational variance Gaussian process regression. Background technique [0002] With the reduction of fossil fuels and the deterioration of environmental pollution levels worldwide, renewable energy has attracted more and more attention. As a clean, huge and recyclable renewable energy, wind energy has become one of the most promising energy sources in the power system. However, due to the intermittent, fluctuating and random nature of wind speed, it is more difficult to connect wind energy to the grid. Accurate and reliable short-term wind speed prediction methods can not only provide guarantee for economical and reliable power generation planning, but also reduce the difficulty of wind power grid integration and improve the efficiency of wind power energy development and utilization. [0003] According to the m...

AI Technical Summary

Problems solved by technology

The heteroscedastic Gaussian process regression can better capture the randomness and volatility of wind speed. However, the maximum a posteriori probability estimation of the heteroscedastic Gaussian process regression model cannot combine all potential independent variables and is prone to overfitting problems.

Images