基于图像特征的锑浮选矿浆ph值预测控制 - 控制与决策.pdf

第31 卷 第11 期 控 制 与 决 策 2016 年 11 月 Vol. 31 No. 11 Control and Decision Nov. 2016 文章编号: 1001-0920 (2016) 11-1973-06 DOI: 10.13195/j.kzyjc.2015.1335 基于图像特征的锑浮选矿浆pH 值预测控制 王晓丽, 曾子骄, 黄 蕾, 谢永芳, 阳春华 ( 中南大学信息科学与工程学院，长沙410083) 摘 要: 适宜的矿浆pH 值是泡沫浮选高效生产的关键. 针对浮选矿浆pH 值无法在线检测和控制滞后的问题, 提取 pH 关联泡沫表面敏感特征, 建立基于仿射传播聚类的多模型最小二乘支持向量机软测量模型, 提出一种基于差分进 化的在线支持向量回归pH 值预测控制方法, 离线建立和在线校正pH 值预测模型, 采用基于差分进化优化方法求解 预测控制决策变量, 从而实现pH 值实时控制. 金锑浮选工业数据表明, 所提出的控制策略稳定了矿浆pH 值, 减少了 药剂消耗. 关键词: 图像特征；差分进化；仿射传播聚类；在线支持向量回归；pH 值预测控制 中图分类号: TP273 文献标志码: A Predictive control of slurry pH based on froth characteristics for antimony flotation process WANG Xiao-li, ZENG Zi-jiao, HUANG Lei, XIE Yong-fang , YANG Chun-hua (School of Information Science and Engineering，Central South University，Changsha 410083，China ．Correspondent: WANG Xiao-li ，E-mail ：xlwang@csu.edu.cn) Abstract: A suitable pH value of the slurry is the key for efficient froth flotation. In the industrial process, it is difficult to measure the pH value online, which causes the control delay. To solve this problem, pH-associated sensitive image features of the froth are obtained, and a soft sensor model-multi-model least squares support vector machine(LSSVM) based on affinity propagation clustering(AP) is introduced. Then, a predictive control strategy based on online support vector regression(OSVR) and differential evolution(DE) optimization for the pH is proposed. The prediction model is built offline and corrected online, and a DE optimization method is used to solve the predictive control problem to find the optimal decisi