A novel hybrid approach based on variational heteroscedastic Gaussian process regression for multi-step ahead wind speed forecasting. (March 2022)

• Record Type: Journal Article
• Authors: Zhang, Chu; Peng, Tian; Nazir, Muhammad Shahzad
• Is Part Of: International journal of electrical power & energy systems. Volume 136(2022)
• Publication Date: 2022-03
• Subjects: Wind speed forecasting -- Heteroscedastic Gaussian processes -- Marginalized variational approximation -- CEEMDAN; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2021.107717
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: VHGPR has good capability in coping with highly nonlinear wind speed signal. VHGPR outperforms the standard GPR in wind speed forecasting accuracy. CEEMDAN can improve the learning ability of the single machine learning models. The proposed CVHGPR model outperforms the other benchmark models. Abstract: Accurate wind speed forecasting is the key to safe and economic operation of electric power and energy systems. As a Bayesian nonparametric method, Gaussian process regression (GPR) has provided competitive forecasting results in recent years. However, conventional GPR model assumes that the noise obeys Gaussian distribution and the variance of the noise in the whole data set is a constant, which is not appropriate for some problems. Motivated by this, this study makes the first attempt to study the ability of the variational heteroscedastic GPR (VHGPR) model in wind speed forecasting. The Marginalized Variational (MV) approximation is employed to approximate the heteroscedastic Gaussian process in the VHGPR model. What's more, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is employed to transfer the nonstationary wind speed series into a certain number of subseries with more regularity such that the forecasting performance of VHGPR can be enhanced. The proposed method is compared with the other twelve benchmark models for 1- to 3- step ahead wind speed forecasting. Experiments results on four real-world datasets demonstrate that VHGPR with the CEEMDAN decomposition strategy is able to obtain better forecasting results for wind speed time series.