浮选加药智能化控制技术研究进展OA

Dosing is the key link to realize the separation and efficient recovery of useful minerals in flotation process.Due to the variability of ore properties and the characteristics of multivariable,nonlinear,strong disturbance and time delay in the flotation process,the traditional dosing control method has obvious deficiencies in dynamic adapt-ability and control accuracy.In recent years,with the rapid development of computer vision,neural network and other technologies and their popularization and application in the field of mineral processing,flotation dosing control has accelerated to intelligent development.This paper summarizes the research status of control system architecture and working principle based on computer vision(CV)state perception and modeling support,intelligent optimization con-trol model construction based on artificial neural network(ANN),intelligent optimization control method based on fuzzy reasoning,intelligent optimization control method based on knowledge reasoning and intelligent optimization con-trol technology of flotation dosing.Constructing a closed-loop control system that integrates multi-modal perception,intelligent reasoning and adaptive optimization capabilities,and combining emerging technologies such as digital twin and reinforcement learning to realize the whole process perception and dynamic optimization of flotation dosing process is an important direction for future research on intelligent optimization control of flotation dosing.