Optimisation of the particle milling process in a vertical roller mill using response surface methodology

In the granular grinding work of the traditional vertical roller mills (VRMs) host, people usually adjust the working parameters in actual use to improve production efficiency and increase service life, but there is no in-depth study of the interaction between the structure and operating parameters. To address the issue, a response surface model and a multi-objective optimisation algorithm were used to optimise and improve the HLM4800 VRM. Box-Behnken response surface experimental design was used to study the grinding disc speed (40-60 rev/min), roll speed (20–30 rev/min), mounting angle (9–15°) and installation clearance (20–60 mm) on particle breaking efficiency and wear rate. The particle breaking process was simulated by employing Rocky-Discrete Element Method (DEM) discrete element simulation software and the Ab-T10 model. A corresponding response surface model was then established to analyze the influencing factors. This approach enabled the identification of the optimal parameter combination for the process. The optimal parameter combination was determined by analyzing the interactions of various factors. The multi-objective optimisation results showed that when the grinding disc speed is 43 rev/min, the roll speed is 21 rev/min, the mounting angle is 14° and the installation clearance is 28 mm, the particle breaking efficiency is highest, increasing by 11%. and the wear rate decreased by 22.82%. Further optimisation scheme simulation and actual operation comparisons showed that the optimisation model prediction error was within 5%, providing data support for the optimisation and upgrade of this type of VRM.