| The science demanding of the computing is increasing along with the developing of science. Because the limit of speed and technology has embarrassed the pc to be used widely, parallel processing is an efficient way to enhance the power of computing. The parallel computing has two class, the special parallel computer and the cluster. The price of the special parallel computer is expensive, and few people can afford, so the application and development of the cluster technology attract more and more people.The Genetic algorithms are a kind of stochastic methods, and its search behavior includes global search and local search. So it has very strong global optimization ability and applied widely in many fields. But the algorithms exists some problems, for example, it can not afford the capacity for efficient if the solved task is very large scale. The parallel genetic algorithm combined the high speed parallel processing and the inherence parallel of the genetic algorithms, improve the computing efficient and also affect the genetic population evolution.Suspension is an important part of vehicle, it greatly impacts vehicle's performances, such as ride comfort, control stabilization, etc. Flexible multi-body vehicle model is a physics model of suspension and designed similar with the real vehicle. The optimization of the parameters of suspension can improve the performation of the run. The deterministic method is easy to fall the search process into local optimization solutions, and is very difficult to achieve exploration.This paper applies the parallel genetic algorithms to optimize the parameters of the vehicle suspension, parallel process the population and calculate the adaptation values. The result indicates that quality and efficient have been improved. Under this condition, we also design the coarse grained parallel algorithm, speedup the individual transfer, and keep the diversity of the population, reduce the early convergence and improve the quality.
|
PDF Full Text Request