Bassey Etim Nyong-Bassey
Energy Management Strategies for hybrid energy systems operation.
- Email: b.e.nyong-bassey1@ncl.ac.uk
Supervisors
- Dr Damian Giaouris
- Dr Haris Patsios
- Prof Simiar Papadopoulou
Project description
Hybrid energy systems operation is often dependent on a pre-defined energy management strategy (EMS). Such a strategy provides optimal decision making.
We can derive EMS for hybrid energy systems using the Power Pinch Analysis within a Model Predictive framework. This necessitates the anticipation of all possible optimal control sequences and corresponding actions. Even so, the EMS is realised under ideal scenarios. It is thus affected by forecast uncertainty.
Q-learning is model-free. It uses stochastic transitions and rewards. We can use such reinforcement learning control techniques to explore and evaluate the possible results of undertaking several control actions for optimal EMS.
In this research, we exploited an adaptive Power Pinch Analysis. This was deterministic and probabilistic. It was within a deep reinforcement learning framework to negate the effects of uncertainty.
Our future work is aimed at integrating demand side response for load shifting (from peak periods to off-peak) into the proposed framework. We will do this via a multi-agent approach to enhance the use of resources.
Publications
- Nyong-Bassey BE, Giaouris D, Patsios H, Gadoue S, Papadopoulos AI, Seferlis P, Voutetakis S, Papadopoulou S. Probabilistic adaptive model predictive power pinch analysis (PoPA) energy management approach to uncertainty. In: The Journal of Engineering 2019, 2019(17), 4288-4292.
- Nyong-Bassey BE, Giaouris D, Papadopoulos AI, Patsios H, Papadopoulou S, Voutetakis S, Serferlis P, Walker S, Taylor P, Gadoue, S. Adaptive Power Pinch Analysis for Energy management of Hybrid Energy Storage Systems. In: IEEE International Symposium on Circuits and Systems (ISCAS) 2018. IEEE.
Qualifications
- MSc Automation and Control
- BEng Electrical and Electronic Engineering
Interests
Deep reinforcement learning, energy systems, probabilistic model predictive control, power pinch analysis..