论文部分内容阅读
磨矿产品粒度直接关系到选矿厂的金属回收率、精矿品位等技术指标,针对磨矿过程滞后时间长、参数时变严重、强非线性、强耦合等特性,采用案例推理技术(CBR)实现磨矿粒度优化。相似度计算是CBR中案例检索的关键环节,直接关系到案例检索的精度。传统采用欧式距离计算相似度的方法,通常假设案例各属性的权重固定且相互独立,而该假设往往不能满足实际应用。针对此问题,提出一种基于粒子群优化算法(PSO)的自学习相似度计算方法,并将其引入案例推理中,构成粒度指标智能优化设定系统,并联合常规的基础控制系统,构建了磨矿过程优化设定控制系统,保证磨矿过程整体优化稳定运行。应用到某大型选矿厂的磨矿流程,取得了明显成效,具有推广应用价值。
The grain size of grinding products is directly related to the technical indicators such as the metal recovery rate and concentrate grade of the concentrator. Aiming at the characteristics of lag time, serious time-varying parameters, strong non-linearity and strong coupling in the grinding process, case-based reasoning (CBR) Achieve grinding particle size optimization. The similarity calculation is the key link of case retrieval in CBR, which is directly related to the accuracy of case retrieval. The traditional method of calculating the similarity using Euclidean distance usually assumes that the weight of each attribute of the case is fixed and independent, and this assumption often can not meet the practical application. In order to solve this problem, this paper proposes a PSO-based self-learning similarity calculation method, which is introduced into case-based reasoning to form an intelligent optimization setting system of granularity index. Combined with the conventional basic control system, Grinding process optimization set control system to ensure that the overall grinding process optimization and stable operation. Applied to the grinding process of a large concentrator, has achieved remarkable results, with the promotion and application value.