Mixed aleatory/epistemic uncertainty analysis and optimization for minimum EEDI hull form design

摘要

EEDI (Energy Efficiency Design Indicator), a mandatory regulation, is closely related to many design parameters of ships' hulls, which were conventionally set to be constant. However, it is often the case that considerable parameter fluctuations occur during actual navigation, so it is more reasonable to state the important parameters as uncertainty variables. There are two primary types of uncertainty: aleatory uncertainty, which exists objectively in architecture and its operating environment, and epistemic uncertainty, deriving from the ignorance or incomplete information of designers. Accordingly, a mixed aleatory/epistemic uncertainty analysis and optimization method based on probability and evidence theory is introduced in this research. Furthermore, transformation methods of uncertainty targets and constraints to certainty targets are employed to conduct a mixed uncertainty optimization. Upon comparison with optimization design of certainty and traditional uncertainty, mixed aleatory/epistemic uncertainty optimization shows excellent adaptability and reliability in minimum EEDI ship hull lines' designs.