LAUSR

The geometry of engineering systems needs to be optimised in the initial design stage. However, engineering system problems consist of multi-objective optimisation and the performance analysis using commercial code is generally time consuming. To optimise the engineering system concerning its performance, many engineers/researchers perform the optimisation using an approximation model. The response surface method (RSM) is usually used to predict the system performance in many research fields, but it shows predic-tion errors for highly nonlinear problems. To create an appropriate response surface model for marine systems, this paper first com-pares the prediction accuracy of the approximation model generated by the RSM, kriging method and artificial neural network (ANN) using a nonlinear mathematical function problem, and optimal design framework is proposed based on a confirmed approximation model. The proposed framework is composed of three parts: definition of geometry, generation of approximation model, and optimisation. The major objective of this paper is to confirm the applicability/usability of the proposed optimal design framework for marine sys-tems. To reduce the time for performance analysis and minimize the prediction errors, the approximation model is generated using the back-propagation neural network (BPNN) which is considered as a neuro-response surface method (NRSM). The optimisation is done for the generated approximation model by non-dominated sorting genetic algorithm-II (NSGA-II). Through derrick structure optimisation design problem considering structural performance and sensitivity analysis, we have confirmed the proposed framework applicability.