Optimal operation of hybrid energy system for intelligent ship: An ultrahigh-dimensional model and control method. (15th November 2020)

• Record Type: Journal Article
• Authors: Tang, Ruoli; An, Qing; Xu, Fan; Zhang, Xiaodi; Li, Xin; Lai, Jingang; Dong, Zhengcheng
• Is Part Of: Energy. Volume 211(2020)
• Publication Date: 2020-11-15
• Subjects: Intelligent ship -- Green port -- Hybrid energy system -- Energy management -- Optimal scheduling; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.119077
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Due to the upcoming regulations for decreasing emissions from shipping, the hybrid energy system is becoming a more popular and feasible option for large ships. This work focuses on the cooperation and optimal operation of onboard hybrid energy system and on-land shore power, in order to obtain cost savings from the perspective of shipping company. Firstly, this work proposes a novel concept, namely "data acquisition and sharing mechanism at shipping company level", then the corresponding operation model and control methodology are developed. To be specific, the ship's energy-management is modelled as an ultrahigh-dimensional optimization problem, in which the binary/continuous hybrid encoding scheme is developed, and a high time-resolution (15 min per point) is achieved. Then, by analysing the variable-coupling characteristic, the proposed model is proved to be non-separable in the early evolution stage, and nearly separable in the later stage. Finally, the novel "power/time/random" variable-grouping and "guided mutation of context-vector" mechanisms are proposed for enhancing the co-evolutionary algorithm to optimize the model. Experimental results show that even when the model dimensionality increases to more than 3000, the evaluated system can strictly satisfy the regulations and constraints from both the port and ship, and also obtain great cost savings. Highlights: Data acquisition and sharing at shipping company level. Ultrahigh-dimensional energy management model for optimal operation of intelligent ship. High time-resolution of 15 min for optimal energy management of intelligent ship. Mathematical proof for variable-coupling characteristic of energy management model. Variable-grouping and mutation mechanisms for co-evolutionary optimization algorithm.