基于子集融合与规则简约的磨矿过程模糊建模
Grinding process modeling based on fuzzy sets merging and rule simplification

针对选矿厂磨矿生产过程的模糊建模问题, 本文提出一种基于模糊集融合和规则简约的模糊建模方法. 该方法针对基于数据建立的磨矿过程Takagi-Sugeno模型, 采用模糊C均值聚类方法对同一变量下的隶属度函数参数进行聚类, 得到对不同工况具有代表性的融合后的隶属度函数, 来降低过度拟合的影响. 此外, 本文根据规则库中的规则权值, 对前件相同的冗余规则进行约简, 形成最终的离线模糊规则库, 有效提高了规则库的泛化能力. 为验证本文方法的有效性, 分别采用经典数据与实际工业数据进行了实验论证, 从精度和泛化能力上体现了本文方法的优势.
In modeling the typical grinding process, we propose a fuzzy modeling method based on fuzzy sets merging and rule simplification. In the fuzzy rule extraction process for the Takagi-Sugeno model, the proposed method adopts fuzzy C mean clustering to partition the fuzzy membership function of every variable to obtain the representative merged membership functions for every working condition to reduce the negative impact of the over-fit phenomenon. Besides, using the weight of each rule, we simplify the fuzzy rulebase by merging the fuzzy rules with the same premises to obtain the final fuzzy model. To validate the proposed approach, a series of comparative experiments are carried out by using classic data and industrial data. The experimental results demonstrate a remarkable generalization ability and practical application potential of the proposed method.