Flotation Concentrate Grade Prediction Model Based on RBF Neural Network & Immune Evolution Algorithm

摘要：

In the process of mineral flotation, the foam in different state represents different concentrate grade. According to this feature, a kind of concentrate grade prediction model (CGPM) was proposed based on the foam image characteristic (FIC). Using RBF neural network based on simulated annealing and fuzzy c-mean clustering algorithm, we established the prediction model between FIC parameter and concentrate grade, and then the model parameters were optimized by immune evolution algorithm (IEA) to improve the model accuracy. The simulation test shows that the model is higher in accuracy and stronger in practicability and robustness, and can give effective guidelines to flotation follow-up dosing control and technical and economic indexes assessment.

关键词： Flotation Foam Image Characteristic RBF IEA

作者: ZHANG Yong JIANG KeJun WANG YuKun

作者单位: School of Electronic and Information Engineering, University of Science and Technology Liaoning, Anshan 114051

会议类型: 国际会议

会议名称: The 31st Chinese Control Conference(第三十一届中国控制会议)

会议地点: 合肥

会议语种:英文

页码: 3319-3323

在线出版日期: 2012-07-01（万方平台首次上网日期，不代表论文的发表时间）