The Improvement Of Genetic Algorithm And Its Application On Carriage Frame Optimization

Genetic algorithm is a random search and optimization method based on natural selection and genetic mechanism of the living beings, which is good at searching global optimum, but its convergence speed is slow;random direction method has fast convergence, but is prone to convergence at local optimum. On the other hand, the mechanical structure optimizations belong to constrained optimization problems with nonlinear equations. For this reason, this paper divides the optimal process into two steps: first, searching the minimum (maximum) solutions with genetic algorithm that adopts the tactic of keeping the best individual: second, searching the optimal solutions with random direction method, the beginning point is the outcomes of the first step.The traditional frame optimization design method has some shortcomings such as too much design circle, repeated work and money wasting. After analyzing the structural characters of carriage frame, the paper introduces a new design method to the strength and stiffness design of carriage frame which solves some problems during optimization design. At the same time, the structural parameters optimization design method of the frame's side-member and cross-member was introduced, and then the objective function and restriction conditions of carriage frame were built up. The mathematical modal was solved with improved genetic algorithm, the result indicated that the improved genetic algorithm-based optimal design could be used to the optimal design problems.