Papers by Keyword: Solution Space

Abstract: We present a novel approach for automatically create industrial products, namely powertrains consisting of engine, transmission and power shaft. We apply a genetic algorithm for exploring the solution space, consisting of 3000 variants, using various criteria, such as power, efficiency and rotation speed. We compare our results with the ones obtained by a human expert in terms of number of feasible solutions, respectively in terms of best and average price. We prove that the proposed approach is generally better than a human expert in exploring the solution space.

Abstract: The particle swarm optimization (PSO) is a population-based stochastic evolutionary algorithm, noted for its capability of searching for the global optimum of complex problems. Particles flying out of the solution space will lead to invalid solutions. So often in engineering applications, boundary condition is used to confine the particles within the solution space. In this paper, a new boundary is proposed, which is called as escape boundary. The solution space is divided into three sections, that is, the inside，escape boundary and the outside of the boundary. The location of the global solution in the solution space, accordingly has two types, that is, the global optimum around the center of the solution space, and the global optimum close to the escape boundary. The proposed boundary is introduced into the PSO algorithm, and is compared to the damping boundary. The experimental results show that the PSO based on escape boundary has better search ability and faster convergence rate.

Abstract: We propose a novel cockroach swarm optimization(CSO) algorithm for Traveling Salesman Problem(TSP) in this paper .In CSO, a series of biological behavior of cockroach are simulated such as grouping living and searching food ,moving-nest, individual equal and so on. For cockroaches crawl and search the optimal solution in the solution space, we assume that the solution which has been searched as the food can split up some new food around solution’s position. The experimental results demonstrate that the CSO has better performance than particle swarm optimization in TSP.