Multi-objective optimization of concrete pumping S-pipe based on DEM and NSGA-II algorithm

In the process of fresh concrete pumping, the S-pipe plays a crucial role in the pumping system. The impact and scratching of concrete aggregate on the wall of S-pipe will cause the wear of the wall and lessen its service life. The structural design of the pumping S-pipe typically relies on experience, and the cost for experiments is too expensive. In addition, the influence mechanism of structural parameters on S-pipe wear is still unclear. In view of this, to reduce the wear of S-pipe and investigate the influence of structural parameters on the wear of S-pipe, the structural parameters of S-pipe (curvature radius r1 and r2, horizontal dimension l3, and inclination angle θ) were optimised through discrete element numerical simulation combined with the non-dominated sorting genetic algorithm-II (NSGA-II) in this paper. Firstly, the wear coefficient was determined by friction and wear test. The discrete element method (DEM) model of pumping suction unit was simplified, and the S-pipe was parameterised. Secondly, sample points and their exact values of the optimization objectives were obtained by Latin hypercube sampling (LHS) plan and DEM simulations, respectively. From these, a Kriging surrogate model was constructed. Taking the average wear rate and the maximum wear of the S-pipe as the optimization objectives, a multi-objective optimization design was performed based on the NSGA-II and the Kriging models. After optimization, a series of Pareto optimal S-pipe models were obtained. Comparing the optimised model with the original model, the average wear rate and maximum wear amount of the S-pipe were reduced by 9% and 26%, respectively.