The Comparative Study And Application To Ship's Double Bottom Structure Of Intelligent Optimization Algorithms

With the development of the shipbuilding industry, the optimization design of ship structure has been a popular study point, and the methods of structural optimization attract a wide attention.In the early optimization study, the structural optimization algorithms based on mathematic, such as gradient method, Newton method, had been used to solve the complex combination optimum problems of engineering. Most of these traditional structural optimizing methods have to use the derivative value of objective function in their optimization searching process. However, owing to most of the objective functions are complicated multivariable and nonlinear functions in the optimization of large scale and complicated structures, the derivative values of objective functions are very hard to obtain, which induce the limitation of the application of the traditional structural optimizing methods. With the development of the artificial intelligence, a lot of intelligent optimization algorithms appeared, such as genetic algorithm, ant colony algorithm and particle swarm optimization algorithm and so on, which provide new idea and method to solve complicated problem of optimization. Verified by large number of solutions, the error probability of this kind of algorithm could be reduced to engineering scope.In this paper, the main research contents are as follows:1. This paper studies and analyzes the modern intelligent optimization algorithms, such as genetic algorithm, ant colony algorithm, particle swarm algorithm. Through experimenting and adjusting parameter repeatedly, a lot of optimization performances could be improved, such as robustness, reliability, universalization and convergence and so on.2. Considering the differences between the discrete domain structural optimization and continuous domain structural optimization, this paper studies the searching process of the modern intelligent optimization algorithms, produces different optimization programs which could be fit for different structural optimization problems, and bring up some suggestions about the technique detail of optimizing, such as algorithms'modeling, parameter selecting an so on.3. Through establishing mathematic optimization model based on the structural rules correctly and exactly, the modern intelligent optimization algorithms are introduced into the optimum design of ship double bottom structure, which shows the advangtages in solving the problem of complex structural optimization.4. In this paper, a new passive heuristic particle swarm optimization (PHPSO) is proposed. The author uses the standard PSO and the PHPSO to optimize the test functions respectively to analyze the PHPSO's performance of optimization comparatively and to set the parameter of PHPSO properly. Finally, the PHPSO is applied in some double bottom structure, and the results show that the PHPSO algorithm is effective in the optimization design of ship structure.