基于改进粒子群算法的机车二系弹簧载荷分配优化

Optimization of locomotive secondary spring loads distribution based on improved particle swarm optimization algorithm

针对机车二系弹簧载荷优化调整这一复杂的多变量优化问题,为进一步提高现有求解方法的优化效果和计算效率,将烟花算法融入粒子群算法,提出一种具有分层递阶结构的改进粒子群算法,算法为3层架构,其中,底层是基础层,为加入维变异算子的粒子群算法,是改进算法的基本框架;中间层是融合层,为引入烟花算法爆炸机制的粒子更新层,主要用于扩大算法搜索范围,提高全局搜索能力;顶层是扰动层,引入扰动因子,避免算法因陷入局部搜索而进行的大量无为冗余迭代,加快全局收敛速度。用典型测试函数对改进算法性能进行测试,并将其应用于机车二系弹簧载荷分配优化调整仿真实验。研究结果表明:改进算法与传统遗传算法、烟花算法和粒子群算法相比,全局搜索能力更强,鲁棒性更好,求解精度更高。

The optimization of spring loads distribution can be summarized as a complicated nonlinear multidimensional optimization problem.In order to further improve the optimization effect and calculation efficiency of the existing solution methods,an improved particle swarm optimization algorithm(IPSO)with hierarchical structure was proposed by integrating the fireworks algorithm(FWA)into a particle swarm optimization algorithm(PSO).The IPSO is a three-layer architecture.The underlying layer was the basic framework of the IPSO and a classical PSO that adopted a dimensional mutation operator.The middle layer was a fusion layer used to expand the search range of IPSO.In essence,it was a particle update layer that introduced the explosion mechanism in FWA.The top layer was the disturbance layer,with a disturbance factor introduced to avoid redundant iteration due to local search and speed up the global convergence.The IPSO was tested on typical test function and applied to the simulation experiment of the distribution optimization and adjustment of the locomotive secondary spring loads.The results show that compared with traditional genetic algorithm(GA),FWA and PSO,the IPSO has stronger global search ability,better robustness and higher accuracy.