Integrated deep learning model for predicting electrical power generation from wave energy converter

Ni, Chenhua and Ma, Xiandong and Wang, Ji (2019) Integrated deep learning model for predicting electrical power generation from wave energy converter. In: 2019 25th IEEE International Conference on Automation & Computing (ICAC'19). IEEE. ISBN 9781728125183

[img]
Text (PID6014839)
PID6014839.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (910kB)

Abstract

The continuous improvement of wave and tidal energy technologies has widely boosted the development of marine energy plants. An accurate predication of the electrical power generation of marine energy not only saves costs for operation and maintenance but also improves manage the electricity consumption and reduce the uncertainty due to the intermittency of the ocean wave and tidal resources. This paper presents an integrated deep learning (DL) network comprising the long short term memory (LSTM) algorithm and the principal component analysis (PCA) to predict the electrical power generation from a wave energy converter (WEC). The results from this integrated data-driven model show the remarkable performance compared with the LSTM alone and other machine learning models. Furthermore, the experiments have shown that the high-frequency oscillating waves and the long term features in the wave have a significant impact on the model’s accuracy. This finding demonstrates the superiority of the proposed model in its ability to deal with time sequence data and the effect of the high-frequency oscillating signals on the results over other machine learning methods.

Item Type:
Contribution in Book/Report/Proceedings
Additional Information:
©2019 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Subjects:
ID Code:
137993
Deposited By:
Deposited On:
21 Oct 2019 11:15
Refereed?:
Yes
Published?:
Published
Last Modified:
24 Nov 2020 11:56