Lightweight Technology Of Ship Structure Based On Surrogate Model

The Lightweight technology,one of the recognized energy-saving and emission reduction measures in the international maritime community,can help to reduce the cost of shipbuilding,improve the maneuverability of ships in navigation,and improve the comprehensive economic performance of ships effectively.In the process of implementing lightweight technology,the surrogate model is more convenient than finite element model in analysis.However,there are many parameters and high degree of non-linearity in the ship structure optimization,so it is imperative to conquer the surrogate model technology with both high accuracy and high computational efficiency.in lightweight technology.The lightweight design in real ships are still concentrating on deterministic optimization,and it ignores some uncertainties such as geometric size,load fluctuation,material properties and artificial operation,which will inevitably lead to the deterioration of the working quality of the structure.Optimal design which considering reliability can provide the guarantee for the reliability of the structure.Therefore,it is necessary to consider the structural reliability in ship structural optimization design.On the basis of previous studies,the surrogate model technology are further studied,and the dynamic surrogate model technology is applied to the lightweight technology of ship structure.At the same time,the structural reliability requirements are taken into account in the process of ship structure optimization to ensure the performance of ship.Based on the theoretical research of surrogate model and experimental design technology,4 numerical examples of non-linear mathematical functions(response surface model,Kriging model,radial basis function model and BP(Back Propagation)neural network model are used to analyze the effects on the prediction accuracy by different number of sample points.Aiming at the problem that optimization of satic surrogate model can’t meet the requirements of convergence and computational efficiency due to lacking global precision,this paper combined trust region and main effect,and proposed a optimization of dynamic surrogate model based on trust region and main effect analysis(ME-TR).The update sampling space is determined by trust region and main effect analysis.The sampling sapce is reduced and updated by trust region strategy,so the local accuracy of the model is improved;The mian effect diagram of variables is obtained by mian effect analysis,and the key area is determined from the diagram,which avoid the shortcoming of trust region that the global optimal solution is easy to miss;Kriging model is chose as surrogate model,which have high gloal accuracy;Particle Swarm Optimization guarantees the efficiency of calculating global optimal solution.Some classical mathematical instances and a instance of ship structure show that the effectiveness and advantage of the proposed strategy.Aimed at the problem that ship structures reliability-based design optimization has poor efficiency and difficult convergence due to high nonlinearity,a reliability-based design optimization of ship structures based on SMOTE(Synthetic Minority Oversampling Technique)algorithm is proposed.The innovation is to establish an improved BP neural network by SMOTE algorithm and the limit state function with high approximation could be set up with fewer samples,which overcomes the shortcoming that old model cannot satisfy both accuracy and efficiency requirements.Moreover,the feasibility of establishing BP neural network model by SMOTE algorithm has been proved by a mathematical instance.Then put the improved BP neural network and simulate anneal arithmetic into single loop approach optimized strategy to solve the issue of ship structure reliability-based design optimization.The result confirms the efficiency and accuracy of the model.In addition,the proposed model can provide an idea for large engineering structures reliability-based design optimization.