Artificial Intelligence Machine Learning in Ship Hydrodynamics

The field of international shipping has undergone significant regulatory changes in 2023, driven by the International Maritime Organization's (IMO) updated strategy for reducing greenhouse gas emissions from ships. The IMO GHG strategy 2023 focuses on enhancing energy efficiency design requirements for new ships and promoting the adoption of zero or near-zero emissions technologies. Additionally, the Energy Efficiency Existing Ship Index (EEXI) and Carbon Intensity Indicator (CII) have become mandatory as of January 1, 2023, urging ship owners and managers to assess and improve their vessels' environmental performance.

Accordingly, DNV Maritime Forecast 2050 (2022) expects that both EEXI and CII will have a significant impact on ship design and operations. To this end, one of the possible solutions to this problem is optimized designs as well as retrofitting ships through improving hydrodynamic performance. To understand the effects of all parameters on the design/retrofit process, digital technologies such as Artificial Intelligence Machine Learning and optimization platforms need to be used and developed (DNV Maritime Forecast 2050).

To address the new and future IMO regulations and in line with corresponding DNV Maritime Forecast 2050, this project will seek two main objectives in parallel. Firstly, it seeks to enhance expertise and competences in the development and implementation of ML methods, as the future computational tools instead of traditional methods, for ship hydrodynamics. It will not only drive advancements in computational tools but also influence the education of future naval architecture students, directly and indirectly. Secondly, the project aims to create a digital design process for ship hydrodynamics, utilizing ML, to support sustainable shipping practices through intelligent ship design. To this end, a Swedish strategic partnership has been built. KTH Royal Institute of Technology, RISE-Research Institutes of Sweden (RISE) and Chalmers University of Technology will collaborate to incorporate ML methods in early-stage ship design, allowing for the exploration of numerous hull shape combinations to identify optimized solutions efficiently.

The digital design process involves investigating new concepts, utilizing various data sources (e.g., experimental data, simulations, and reduced order models), and integrating ML techniques with multi-disciplinary design optimization. The overarching goal is to reduce computational costs and optimize marine vehicle design. Moreover, the project emphasizes collaboration among designers, developers of computational tools, and experts in ML and multi-disciplinary design optimization to achieve optimal results. An expert reference group, including the Swedish Defence Research Agency (FOI), Stena, the Swedish Transport Agency, the Swedish Defence Materiel Administration (FMV), Sjöbefälsföreningen, SALTECH Consultants, ScandiNAOS, FKAB Marine Design, and Wallenious Marine, supports the project, showcasing a commitment to cooperation across different domains.