Optimization of the Dewatering Process for Concentrate Pressure Filtering by Support Vector Regression

Abstract

In order to better improve the efficiency of the concentrate filter press dehydration operation, this paper studies the mechanism and optimization methods of the filter press dehydration process. Machine learning models of RBF-OLS, RBF-GRNN and support vector regression (SVR) are constructed respectively, and Perform laboratory simulation and industrial simulation separately. SVR achieves the best accuracy in industrial simulation, the simulated mean relative error (MRE) of moisture and processing capacity are respectively 1.57% and 3.81%. Finally, a simulation model of the filter press dehydration process established by SVR, and the optimtical simulation results Obtained by optimization method based on control variables. The results show that the machine learning method of SVR and optimization methods based on control variables are applied to industry, which can not only ensure the stability of expected production indicators, but also shorten the filter press dehydration cycle to less than 85% of the original.