Collaborative optimization of grinding-roll operating and flute-profile parameters

The roller mill is critical in wheat milling, where grinding-roll operating and flute-profile parameters affect efficiency, energy consumption, and lifespan. While previous studies often address these factors independently, research on their synergistic optimisation is limited. This paper employs the Discrete Element Method (DEM) to construct a grinding model for roll and wheat grains. Response Surface Methodology (RSM) is used to establish second-order polynomial regression models linking five key parameters—roll gap, speed ratio, number of flutes, width of land, and front angle—to grinding power and roll wear depth, with all models achieving R2, Adjusted R2, and Predicted R2 values exceeding 0.98, indicating excellent fit. Interactive effects of the parameters are analyzed,and NSGA-II is employed to synergistically optimisation grinding power and wear values, generating 95 sets of Pareto optimal solutions. The entropy-weighted TOPSIS method identifies the optimal combination:: roll gap 1.0147 mm, speed ratio 1.25, number of flutes 3.4258 flutes/cm, width of land 0.2233 mm, and front angle 64.5811°. Validation shows this reduced grinding power by 0.2213 kW, decreased fast roll wear depth by 1.5258 × 10−9 mm, and lowered slow roll wear depth by 6.104 × 10−10 mm. This effectively improves energy efficiency and service life, and provides a methodological foundation for the systematic optimisation of grinding-roll parameters.