基于克里金和粒子群算法的装载机铲掘轨迹规划

Shoveling trajectory planning method for wheel loader based on kriging and particle swarm optimization

以某50型轮式装载机为研究对象,结合其行驶、作业特性和自身结构特点,联合RecurDyn和EDEM建立了包含物料信息的装载机铲掘特性仿真模型。归纳驾驶员经验分析了装载机铲掘散状物料的过程,构建了能反映铲斗满斗率和燃油消耗量综合性能的装载机铲掘性能指标。针对不同工况和不同铲掘轨迹参数下的装载机铲掘性能,基于Kriging方法建立了不同工况下铲掘性能关于铲掘轨迹参数的近似响应关系,采用粒子群算法对其进行优化,得到了不同工况下的理论铲掘轨迹参数。分析结果表明:优化后的铲掘轨迹参数所对应的装载机铲掘性能将会得到明显提升。

This paper took a 50-type wheel loader as the research object. Considering the driving,working and structural characteristics of the wheel loader,a loader simulation model containing the loaded material information was established using RecurDyn and EDEM. Based on the driver′s experience,the process of shoveling bulk materials was analyzed,and the shovel performance indexes reflecting the comprehensive performance of the bucket filling rate and fuel consumption were constructed. On the above basis,the shovel performance under different working conditions and different trajectory parameters was analyzed. The approximate response relationship between the shovel performance and the trajectory parameters under different working conditions was established using Kriging method. The theoretical trajectory parameters of the shovel under different working conditions were obtained by using Particle Swarm Optimization(PSO). The results show that the shovel performance is significantly improved with the optimized trajectory parameters.