某矿用车平顺性优化仿真与试验研究

发布时间：2018-10-12 19:59

【摘要】：为了解决矿用人车平顺性较差的问题，结合车辆平顺性试验方法，将改进粒子群算法、灵敏度分析方法和近似建模理论应用到求解悬架多参数优化问题当中，并通过试验验证，提出一种有效的悬架多参数优化方法。本文按照从低自由度到高自由度建模，从简单到复杂分析问题的思路展开。首先基于1/2车辆振动模型，以车身振动加权加速度均方根为目标函数，以悬架动挠度和轮胎动载荷为约束，运用标准粒子群算法（SPSO）进行车辆四自由度悬架参数优化。针对SPSO算法容易陷入局部最优、优化速度较慢的问题，通过惯性权值调整、超出边界粒子速度位置选择、引入混沌变异加强局部搜索和调整混沌粒子群算法邻域选取策略等方法提高了粒子群算法寻优精度和收敛速度，并提出了指数函数调整惯性权值和局部邻域的混沌粒子群算法（ICPSO）。仿真表明：应用ICPSO算法可以大大提高悬架优化问题的收敛速度和寻优精度，获得最优的悬架参数匹配结果。将改进粒子群算法应用到整车七自由度振动模型中，针对粒子群算法多次调用模型求解运算效率低、耗时长的问题，对影响车辆平顺性的18个悬架主要参数进行灵敏度分析，找出对车辆平顺性影响较大的参数，，并对车辆振动模型进行响应面法近似建模，用二阶多项式拟合仿真模型，最后再用ICPSO算法对悬架近似模型进行优化。仿真表明：应用近似模型大大减少了优化时间，提高了优化效率，得到了理想的悬架参数匹配结果，在悬架动挠度变化不大的情况下，大大降低了车身振动加权加速度均方根值和轮胎动载荷。最后参照GB/T4970-2009平顺性试验方法对改进前和改进后的车辆进行了随机和脉冲路面平顺性试验，试验结果表明：改进后的车辆随机路面加权加速度均方根值比改进前减小了30%左右，脉冲路面最大加速度响应值比改进前减小了50%左右，试验结果证明了基于粒子群算法的悬架多参数优化可以提高车辆的行驶平顺性，可以用来指导后续的悬架开发和设计。
[Abstract]:In order to solve the problem of poor ride comfort of mine vehicles, the improved particle swarm optimization (PSO) algorithm, sensitivity analysis method and approximate modeling theory are applied to solve the multi-parameter optimization problem of suspension. An effective multi-parameter optimization method for suspension is proposed. In this paper, the idea of modeling from low degree of freedom to high degree of freedom and from simple to complex analysis is presented. Firstly, based on 1 / 2 vehicle vibration model, taking the root-mean-square root of vehicle body vibration weighted acceleration as objective function, taking suspension dynamic deflection and tire dynamic load as constraints, the standard particle swarm optimization algorithm (SPSO) is used to optimize vehicle four-degree-of-freedom suspension parameters. Aiming at the problem that SPSO algorithm is easy to fall into local optimum and the speed of optimization is slow, the velocity position of particles beyond the boundary is chosen by adjusting the inertia weight. By introducing chaos mutation to enhance local search and adjust the neighborhood selection strategy of chaotic particle swarm optimization algorithm, the optimization accuracy and convergence speed of particle swarm optimization algorithm are improved, and a chaotic particle swarm optimization algorithm, (ICPSO)., which adjusts inertia weight and local neighborhood by exponential function is proposed. Simulation results show that the convergence speed and precision of suspension optimization problem can be greatly improved by using ICPSO algorithm, and the optimal suspension parameter matching results can be obtained. The improved particle swarm optimization (PSO) algorithm is applied to the vibration model of vehicle with seven degrees of freedom. Aiming at the problems of low efficiency and long time consuming, the main parameters of 18 suspensions which affect the ride comfort of the vehicle are analyzed. The parameters which have great influence on vehicle ride comfort are found, and the response surface method is used to model the vehicle vibration model, the simulation model is fitted with second-order polynomial, and the suspension approximate model is optimized by ICPSO algorithm. The simulation results show that the application of the approximate model can greatly reduce the optimization time, improve the optimization efficiency, and obtain the ideal suspension parameter matching results. When the dynamic deflection of the suspension does not change much, the dynamic deflection of the suspension is not changed. The root-mean-square value of vehicle vibration weighted acceleration and the dynamic load of tire are greatly reduced. Finally, the random and pulse pavement ride comfort tests of vehicles before and after improvement are carried out by referring to the GB/T4970-2009 ride comfort test method. The experimental results show that the RMS value of the improved vehicle random pavement weighted acceleration is reduced by about 30% compared with that before the improvement. The maximum acceleration response of impulse pavement is reduced by about 50% compared with that before improvement. The experimental results show that the multi-parameter optimization of suspension based on particle swarm optimization can improve the ride comfort of vehicles and can be used to guide the subsequent suspension development and design.
【学位授予单位】：北京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD50

【参考文献】