关键词: 煤泥浮选, 自动加药, 最小二乘支持向量机, 粒子群优化算法

Abstract: In the coal slime flotation process, manual dosing leads to the problems of strong randomness, high reagent consumption and high labor intensity of workers. In this paper, a new method based on least square support vector machine (LS-SVM) for the predictive control of flotation reagents addition is proposed. The main process variables are selected as the input variables of the LS-SVM model, the reagents dosage are the output variables. Meanwhile, the particle swarm optimization algorithm (PSO) is used to optimize the internal parameters of the model and the simulation is carried out. The application on the Hong tong coal preparation plant shows that: under the premise of ensuring the product quality, the system can effectively reduce the reagent consumption, in which the collector consumption is reduced by 13.73%, the frother consumption is reduced by 12.67%, the model shows good performance.