Browse by author
Lookup NU author(s): Professor Damian Giaouris
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
© 2022 Elsevier LtdHybrid energy storage systems involve the integration of multifarious energy storage technologies which have complementary operational characteristics. Although the incorporation of hybrid energy storage systems can suppress renewable energy intermittency and improve energy supply; control and coordination become more challenging. By integrating more assets and more functions in the system, the complexity becomes prohibiting and hence new versatile and scalable methods must be employed. The authors have in the past proposed a graphical energy management strategy based on the power pinch analysis framework to address this challenge but was contingent on average energy demand and supply profile that reflects only a single scenario case study. This paper proposes a probabilistic adaptive power pinch analysis paradigm for energy management of isolated hybrid energy storage systems with uncertainty, which is implemented in a model predictive receding horizon. The proposed method uses multistate stochastic power grand composite curves realised from the integration of distinct energy demand and supply profiles randomly sampled from historic data via Monte Carlo simulation. Thus, in a predictive horizon, the multiple possibilities which the state of the energy storage can attain must jointly satisfy a probabilistic chance constraint factor with necessary decisions inferred in the control horizon to negate uncertainty. The proposed probabilistic method was further improved by a correction mechanism that minimises the mean squared error between the actual and predicted state of charge of the battery. The performance of probabilistic power pinch analysis was tested on a hybrid energy storage system with renewable energy sources, a battery, a fuel cell and an electrolyser. The results clearly demonstrate improved robustness to Gaussian uncertainty than a day ahead power pinch analysis reference as over-dis/charging the battery and carbon emission were reduced by 98 %, 22 % and 100 % respectively but, necessitates allocating more hydrogen resources. A battery degradation model was also used to evaluate the performance of the probabilistic power pinch analysis energy management strategies.
Author(s): Etim N-BB, Giaouris D
Publication type: Article
Publication status: Published
Journal: Journal of Energy Storage
Year: 2022
Volume: 54
Print publication date: 01/10/2022
Online publication date: 18/07/2022
Acceptance date: 25/06/2022
ISSN (electronic): 2352-152X
Publisher: Elsevier Ltd
URL: https://doi.org/10.1016/j.est.2022.105224
DOI: 10.1016/j.est.2022.105224
Altmetrics provided by Altmetric
