Research On Optimum Design Of Large Crude Oil Tanker Structures

As the main tool of crude oil transportation on the sea, the large crude oil tanker has complex structures and enormous wastage of steel, so the structural optimization of this kind of ship is fairly significant in decrease of steel wastage and improvement of design level. This paper researches on optimum design of large crude oil tanker structures with mathematical programming, aiming at the minimum weight of the cargo tank amidships.The paper optimizes the midship section structure of large crude oil tanker based on DNV rules. The optimization model is built, which has the function of decreasing the number of design variables and improving the efficiency of optimization. The Relative Difference Quotient Algorithm (RDQA) is advanced for the optimum design. The applicable model of Genetic Algorithm for the optimization is built. Results of examples of a 150,000 DWT LCC and a 299,000 DWT VLCC show that the value of objective function decreases by 5.33% and 6.46% respectively.After optimization of the midship section structure, the paper researches on optimum design of large crude oil tanker transverse bulkhead structure based on DNV rules. The high-efficiency optimization project is established and Mutative Scale Chaos Algorithm is introduced. Results of optimization of the 150,000 DWT LCC and 299,000 DWT VLCC show that the weight of transverse bulkhead decreases by 4.86% and 4.13% respectively.An initial research on structural optimization of holistic cargo tank of large crude oil tanker based on direct strength calculation is accomplished in this paper. By using APDL of ANSYS, a general half-one-half cargo tank Finite Element Method (FEM) model, which is the platform of FEM analysis of optimization, is built for cargo tank strength assessment of large crude oil tank with two oil-tight longitudinal bulkheads and a cross tie arrangement in the wing cargo tanks. The optimization model and its interface with ANSYS are built. By using theory of Uniform Design (UD), the paper creates the characteristic swatch sets, and analyzes these experiment points with FEM. Then the RBF (Radial Basic Function) neural network response surface, which has capability to simulate the relations between design variables and stresses of controlling points, is created. Lastly, with the optimization algorithm-RDQA, the optimization project of the cargo tank structures is established. Results of optimization of the 299,000 DWT VLCC show that weight of the cargo tank decreases by 1.93%, indicating the feasibility of the project.