Multi-objective optimization of shearer drum cutting parameters based on DEM and a hybrid NSGA-II/DE algorithm considering productivity, energy, and wear

Balancing productivity, energy consumption, and wear is critical for mining gangue-containing coal seams. This study proposes an integrated framework combining the discrete element method (DEM), response surface methodology (RSM), and a hybrid NSGA-II/DE algorithm to characterize shearer drum wear and optimize cutting parameters. Three tasks were performed: (1) developing a DEM model of a coal wall containing gangue based on mechanical testing and parameter calibration; (2) analyzing pick and blade wear distributions and mechanisms using Archard and relative wear theories, while establishing wear-depth prediction models through RSM; and (3) constructing a multi-objective model considering productivity, specific cutting energy, and drum wear, and solving this model using the hybrid algorithm. Results show that abrasive wear dominates impact wear during drum cutting. Optimization increased productivity by 2.42%, reduced specific cutting energy by 14.29%, and decreased pick and blade wear depths by 8.07% and 1.23%, supporting greener, more efficient, and lower-wear shearer operation.