基于改进PSO-GA算法的轨道精调优化研究

Track fine adjustment optimization based on improved PSO-GA Algorithm

为了解决轨道精调过程中调整量过大且调整效率低的问题,本文提出一种改进粒子群-遗传算法进行调整量优化改善轨道不平顺。该算法充分利用粒子群算法搜索速度快及遗传算法搜索范围广的优点,且在遗传操作中引入最优保存策略,一种新的自适应交叉方式、自适应变异。相较于传统算法,改进后的算法有更强的跳出局部最优、保持活力的能力。研究结果表明:改进PSO-GA算法的实验结果,相较于遗传算法和粒子群算法,调整量的平均值改善了16.1%和5.5%。并且经过该算法调整后的各指标平顺性都优于其它算法,即该算法在可以保证最小调整量,减少工作量的同时又可以确保轨道高平顺性。

In order to solve the problem of the large adjustment and the low efficiency in track fine adjustment process,an improved particle swarm optimization-genetic algorithm is proposed,which makes full use of the advantages of fast searching speed of particle swarm optimization algorithm and wide searching range of genetic algorithm.The optimal saving strategy,the new crossover method and the adaptive mutation is introduced in the genetic operation.Compared with the traditional algorithm,the improved algorithm has stronger ability to jump out of local optimum and maintain vitality.The results show that adjusted amount obtained by the improved algorithm improves the average value of objective function by 16.1%and 5.5%compared with genetic algorithm and particle swarm optimization algorithm.The minimum adjustment amount and the high ride comfort could be ensured and also the workload could be reduced by the algorithm.