Title: A novel online kernel ridge to forecast next-day electricity price

Authors: Jiancheng Yang; Weiwu Yan; Renchao Xu; Xi Zhang

Addresses: Department of Automation, Shanghai Jiao Tong University, Shanghai 200240, China ' Department of Automation, Shanghai Jiao Tong University, Shanghai 200240, China; Key Laboratory of System Control and Information Processing, Ministry of Education, Shanghai, China ' Department of Automation, Shanghai Jiao Tong University, Shanghai 200240, China ' China Southern Power Grid International Co., Ltd., Guangzhou 510623, China

Abstract: Accurate prediction for electricity price plays a great role in developing bidding strategy in the competitive energy market. In this brief, a novel variant of kernel ridge, namely multivariate slide-window online kernel ridge, is proposed to capture the nonlinearity and non-stationarity of electricity price as seasonal time series, by handling the timestamps in one period synchronously. Compared to traditional time series techniques like autoregressive integrated moving average (ARIMA) and other techniques like random forest and support vector regression, it provides much higher accuracy with lower computation cost, and can be easily integrated with other related data. Results from EPEX France spot market are presented.

Keywords: online kernel ridge; seasonal time series; electricity market; EPEX France.

DOI: 10.1504/IJSCIP.2018.097198

International Journal of System Control and Information Processing, 2018 Vol.2 No.4, pp.317 - 331

Received: 09 Apr 2018
Accepted: 08 Nov 2018

Published online: 23 Dec 2018 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article