Robust optimization method for co-simulation of equipment based on EDEM-ADAMS

Abstract

Tunneling machines, pivotal in rock tunnel excavation, utilize cutting mechanisms for rock fragmentation. As the core component of the cutting mechanism, the cutting head experiences severe vibrations during the rock breaking process when subjected to large loads, which adversely affects the working performance of the tunneling machines. The precision and efficiency of cutting force simulation for the cutting head are crucial for equipment design optimization and performance assessment. Therefore, exploring robust simulation time steps is particularly significant. This paper leverages state-of-the-art simulation techniques to boost the accuracy and computational performance of cutting head simulation. Firstly, by setting 44 different combinations of EDEM-ADAMS time steps, simulations are conducted in four different environments to collect cutting forces and simulation time data. Then, in view of this dataset, the radial basis function (RBF) approximation model is developed to simultaneously predict cutting forces and simulation time under four environments, which enhances the accuracy and applicability of the predictions. Finally, targeting the minimization of relative error, fluctuation magnitude, and simulation time, the NSGA-II algorithm is further utilized for multi-objective iterative optimization to obtain the time step combination with excellent performance. The results demonstrate that the optimized method reduces the relative error by 67.8 %, the fluctuation magnitude by 43.6 %, and the simulation time by 31.8 %. These improvements highlight the effectiveness of the optimization approach in enhancing both the precision of cutting force prediction and the stability of the simulation process, while maintaining computational efficiency.