Short-term power load forecasting of GWO-KELM based on Kalman filter
Xiaoyu Chen, Yulin Wang, Jianyong Tuo
Abstract
Short-term power load forecasting plays a significant role in power system security management. The prediction model in this paper is the grey wolf optimization algorithm to optimize kernel extreme learning machine (GWO-KELM). First, the Kalman filter is used to reduce the noise for the random noise interference existing in the power load data. Then determine the input and output of the prediction model. In this paper, the ELM model of three different kernel functions is used for comparative experiments, and the mean absolute percentage error is used as the evaluation model index. It is concluded from the experimental results that the GWO-KELM model used in this article has the advantages of high prediction accuracy and strong generalization ability, so it is practicable to apply the model to short-term electric load forecasting.