Knowledge Based Engineering (KBE) automatic layout generation framework for modular offshore wind service vessels

Abstract

The urgent development of green energy has become a worldwide trend in the war against the threat of global warming and the impact of extreme weather. The offshore wind farm is one of the solutions that are able to harvest energy sustainably from the environment. To erect these offshore windfarms, the debate about whether a new fleet should be designed and built or make the existed offshore fleet go under retrofitting has been raised. The stakeholder group involved in this issue is formed by the ship designers, ship operators, and market analysts. The difference between the existed offshore support vessel fleet and the offshore wind farm support fleet is the former is built to perform a certain operation and will require another considerable investment to be retrofitted while the latter is looking for solutions to switch from different equipment thus to keep the flexibility between different operation.
The main objective of this research is to propose a ship design methodology to enable the offshore wind farm's request in mission flexibility by improving the design process in the preliminary design phase. Modularity and Knowledge Based Engineering (KBE) are chosen to be the two main topics to support the research objective. To narrow down the research scope, the research object is limited to a small Offshore support Vessel fleet that is able to perform service throughout the lifecycle of an offshore windfarm.
The application of the modularity concept is to break down the vessel system into smaller subsystem or function blocks then reassemble them to be self-sufficient modules. Modules are the basic elements for forming a modular platform to provide a basic model that is able to perform various operations by mounting different equipment. In the former research done in NTNU, a methodology for manipulating the modularity to assemble a product platform by processing previous configuration scripts has been developed. It has greatly reduced the repetitive and iterative works in the design of similar vessels, especially the design for OSVs. Though the configuration scenarios are flexible to change, the pre-designed modules' properties are limited by the based configuration. On the other hand, the configurations are closely tight to the hull shape thus limited the freedom from both the hull design and configuration design. In this research, an improved configuration generator is developed to separate the configuration design and the hull design. The proposed methodology will also keep the flexibility in importing new configurations and the latest hull shapes without conflicts.
Knowledge based engineering is another topic included in this research. It has been proved and widely used in the aerospace industry. As for the ship design industry, the application is limited to local structure design. KBE is a bridge to bring the user and the designer together in the design project. It consists of a programming stage, Knowledge Based System (KBS), to process the selection of suitable design cases and a visualization stage, Computer-Aided Design (CAD) ,to give a direct review on the whole design project. The fundamental drivers for each KBE system are called "High Level Primitive" (HLP). They are the basic units for the knowledge stored in the database. KBS picks out suitable HLPs from the database and assigned them in a configuration script thus forming a design result stored in the digital world. CAD will take over the result and visualize it in the drawing window. This is the proposed methodology to fill the gap between the modularity re-configuration and the diversity of modules.
A Multi-Model-Generator (MMG) will become the final output of this research. It is designed to be able to reproduce several vessels in the OSV market. A further demonstration of the ability to generate a modular OSV product family at the end of the research. Though the modeling output has been proved to meet the index set for the research object, the MMG has the potential to be improved in the future by importing a well-constructed database from the industry.