A Comprehensive Review of Modeling and Optimization Methods for Ship Energy Systems

Foivos Mylonopoulos*, Henk Polinder, Andrea Coraddu

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)
167 Downloads (Pure)

Abstract

This paper presents a comprehensive literature review of the state-of-the art modeling and optimization methods for the power and propulsion systems of ships. Modeling is a tool to investigate the performance of actual systems by running simulations in the virtual world. There are two main approaches in modeling: physics-based and data-driven, which are both covered in detail in this survey paper. The output from the simulations might not be optimal in terms of certain performance criteria such as energy consumption, fuel cost etc. Hence, it is vital to optimize the systems considering the efficient interaction between the components, to yield the optimal performance for the integrated vessel's powertrain. In this paper, the optimization case studies, for the ship energy systems, will be divided in terms of a) optimal design (topology and sizing), b) optimal control and energy management strategies, c) combined optimal design and control. Tables that summarize the literature review outcomes will also be presented at the end of each section. The main outcome is that limited literature is available for optimizations of ship powertrains using data-driven models, especially surrogate models. Surrogate-assisted optimizations for integrated ship energy systems can yield optimal solutions at fast computational speeds, with sufficient accuracy, even for complex, nested, multi-level, multi-objective optimizations.

Original languageEnglish
Pages (from-to)32697-32707
JournalIEEE Access
Volume11
DOIs
Publication statusPublished - 2023

Keywords

  • Energy systems
  • modeling
  • optimizations
  • ships

Fingerprint

Dive into the research topics of 'A Comprehensive Review of Modeling and Optimization Methods for Ship Energy Systems'. Together they form a unique fingerprint.

Cite this