Short Term Wind Power Prediction Based on Improved Kriging Interpolation, Empirical Mode Decomposition, and Closed-Loop Forecasting Engine

Nima Amjady,Oveis Abedinia

doi:10.3390/su9112104

Abstract

The growing trend of wind generation in power systems and its uncertain nature have recently highlighted the importance of wind power prediction. In this paper a new wind power prediction approach is proposed which includes an improved version of Kriging Interpolation Method (KIM), Empirical Mode Decomposition (EMD), an information-theoretic feature selection method, and a closed-loop forecasting engine. In the proposed approach, EMD decomposes volatile wind power time series into more smooth and well-behaved components. To enhance the performance of EMD, Improved KIM (IKIM) is used instead of Cubic Spline (CS) fitting in it. The proposed IKIM includes the von Karman covariance model whose settings are optimized based on error variance minimization using an evolutionary algorithm. Each component obtained by this EMD decomposition is separately predicted by a closed-loop neural network-based forecasting engine whose inputs are determined by an information-theoretic feature selection method. Wind power prediction results are obtained by combining all individual forecasts of these components. The proposed wind power forecast approach is tested on the real-world wind farms in Spain and Alberta, Canada. The results obtained from the proposed approach are extensively compared with the results of many other wind power prediction methods.

Highlights

Wind power is one of the fastest growing electricity sources in the world [1]
In [9], wind power forecast by Ridgelet Neural Network (RNN) has been proposed in which Ridgelet is used as the activation function of the hidden nodes
A hybrid forecasting method based on Enhanced Particle Swarm Optimization (EPSO) has been introduced in [10] for wind power prediction

Summary

Introduction

Wind power is one of the fastest growing electricity sources in the world [1]. Despite wind power’s clean benefits, wind power is a non-dispatchable resource which is dependent on weather conditions [2]. (4) A new wind power prediction approach is presented, which is composed of the proposed EMD, an information-theoretic feature selection method, and the proposed forecasting engine. The proposed approach decomposes this wind power time series by means of IKIM and EMD, selects most effective features for it by means of Maximum Relevancy, Minimum Redundancy and Maximum Synergy (MRMRMS) feature selection method and provides wind power forecast for the microgrid through the closed-loop forecasting engine. Since wind speed prediction results in terms of Root Mean Square Error (RMSE) have been presented in [8], this forecast process and error criterion have been considered for all methods of Table 1.

Alberta Test Case

Blue Canyon Wind Farm