The model, design, and management of renewable energy harvesting systems for maritime robots

Abstract

Renewable energy harvesting systems could provide sustainable power to supply mobile robots for fully autonomous operation. However, the lack of model theory and design technique in current research limit the renewable energy system to the stationary application only. This thesis proposes a methodology to model, design and manage the power system for maritime robots. A novel non-stationary model including both global and local motion are developed for renewable energy harvesting systems. This offers a new methodology to simulate the power system with a spatial-temporal power generation and demand load model. The model is utilised by a data-driven approach to design a reliable and economic renewable energy system configuration under the size constraint. This design method is based on the optimisation approach to size the configuration of renewable energy devices. Furthermore, a learning-based power management strategy was proposed to improve the long term robustness of the power system. Variation and uncertainty in the renewable energy resource are mitigated by learning to plan the power usage from experience. The non-stationary renewable energy harvesting model shows both global and local motion of the robot influence the power generation and demand load. To match the power generation and demand load of the robot, the size of the renewable energy harvester and storage device has to be optimally designed under strict size constraints. Design optimisation result demonstrated the optimality and feasibility of the proposed renewable energy harvesting system configuration method. For the long-term performance, the learning-based power management strategy outperforms all benchmarking strategies in providing a guaranteed minimum level of power supply. Research results show the renewable energy could provide sustainable power supply to the robot. These methods and techniques provide a foundation to the model, design and management of renewable energy harvesting systems for mobile maritime robots.

More information

Identifiers

ORCID for Yu Cao: orcid.org/0000-0001-5767-7029

ORCID for Nicholas Townsend: orcid.org/0000-0001-6996-3532

Catalogue record

Export record