脉动式无级变速器的混沌粒子群优化设计

针对脉动式无级变速器输出转速脉动度过高的问题,提出一种基于混沌粒子群算法的尺寸优化方法。首先提出一种混沌粒子群算法,即用混沌变异替代原速度和认知部分并增加惩罚项用于求解约束优化问题。其次,以尺寸参数为决策变量,以脉动度为评价指标,结合约束条件建立变速器的适应度函数。最后,应用混沌粒子群算法求解变速器尺寸参数最优解。结果表明,相对于基本粒子群算法,混沌粒子群算法精度和鲁棒性提高一个数量级,求解时间减少70%,优化后的变速器脉动度下降了41%。

Optimal Design of Pulse CVT Based on the Chaos PSO

In order to solve the problem that the output speed of the Pulse CVT is too high,an optimization of Pulse CVT based on the Chaos PSO is proposed. Firstly,the Chaos PSO is proposed which uses chaotic mutation to replace the original velocity and the cognitive part,and adds penalty term to solve the constrained optimization problem. Secondly,taking the dimension parameter as the decision variable andtaking impulsion rate as the evaluation index,the fitness function of the transmission is established with the constraint condition. Finally,the Chaos PSO is applied to solve the optimal parameters of transmission parameters. The results showed that the accuracy and robustnessof the Chaos PSO wasgreatly improved,the solution time was reduced by 70%,and the impulsion rate was decreased by 41%.