Improved Artificial Fish Swarm Algorithm And Its Application In Truss Structure Optimization

2011-2015is the implementation of China’s national economy and social development of12th Five-Year Plan period, it is a critical period of building a moderately prosperous society, it is to deepen reform and opening up. accelerate the transformation of economic development in the crucial period, and at the same time achieving the prosperous period of the construction industry. Truss structure has good capacity, light weight, good coordination and strong seismic ability to withstand the loads from multiple directions, easy accessibility, low cost, transport facilitation, etc., so it is widely used in large span workshop, exhibition halls, gymnasium and bridges and other public buildings. More and more attention should be paid to the truss structure optimization problem in the field of engineering design.In recent years, bionic optimization algorithm to optimize the design of engineering structures began to rise with the development of computer technology, new ideas and means were built to solve structural optimization problems. Artificial fish swarm algorithm (the AFSA) as a new intelligent optimization algorithm was proposed by Li Xiao lei in2002. First, the local optimization was got by the individual fish in artificial fish feeding, clusters and rear-end behavior, then through constant iteration to find the optimal solution of the problem. The algorithm has good global search ability and has the advantages of initial value and parameter selection is sensitive, robust. simple, easy to realize. We try to apply the artificial fish swarm algorithm for optimization of truss structure, in order to provide a new optimization method. The basic artificial fish swarm algorithm was improved, according to the basic artificial fish swarm algorithm convergence speed is slow. optimizing the accuracy is not high and the character of truss structure. At the initialization stage of algorithm, the transverse and randomness of chaos was introduced to initialize fish group. so that the efficiency and quality of solution could be improved. During the running time, the inertia weight adjustment strategy of particle swarm optimization was used to adjust the step of artificial fish, then the speed and accuracy of optimization could be raised. The constraints separation comparison method was introduced to the optimization of improved artificial fish swarm algorithm. then try to apply the improved algorithm to the continuous variable size optimization of truss structure. The concepts of distance. neighborhood, central location of the artificial fish swarm algorithm were improved. The improved algorithm was applied to the discrete variable size optimization of truss. The size of the truss cross-section is regarded as design variables, the minimum structural weight is regarded as the objective function. The results were compared with other algorithms and showed the feasibility and effectiveness of the improved algorithm.