Abstract: Aiming at the dual-layer nonlinear implicit optimization problem of shoveling performance of wheel loader in dynamic uncertain environment,taking the segmented shoveling ore particle pile as a typical working condition,EDEM-ADAMS co-simulation was used to obtain the influencing factors of shoveling performance,and the hyperparameters of Kriging model were optimized by differential evolution cuckoos algorithm.An improved Kriging model for shovel performance of wheel loaders was constructed.On this basis,with the maximum full bucket rate as the upper goal and the minimum shovel resistance as the lower goal,a dual-layer optimization model of wheel loader shovel performance was established,and the whale algorithm was used to solve the problem.After optimization,the full bucket rate was increased by 28.8%,and the shovel resistance was reduced by 10.5%.Compared with the traditional multi-objective optimization method,the shovel resistance was reduced by 17%,and the full bucket rate was increased by 6.5%,which verified the effectiveness and advancement of the proposed method.

Key words: wheel loader, excavating performance, Kriging model, hyperparameters, dual-layer optimization

摘要： 针对动态不确定环境下轮式装载机铲掘性能双层非线性隐式优化问题,以分段式铲掘矿石颗粒料堆为典型工况,采用EDEM-ADAMS联合仿真获取铲掘性能影响因子,通过差分进化布谷鸟算法优化Kriging模型的超参数,构建轮式装载机铲掘性能的改进Kriging模型。在此基础上,以最大满斗率为上层目标、最小铲掘阻力为下层目标,建立轮式装载机铲掘性能双层优化模型,并采用鲸鱼算法进行求解。优化后满斗率提升了28.8%,铲掘阻力降低了10.5%。与传统多目标优化方法相比,铲掘阻力降低了17%,满斗率提高了6.5%,从而验证了所提方法的有效性和先进性。

关键词: 轮式装载机, 铲掘性能, Kriging模型, 超参数, 双层优化