摘要
基于混沌优化的思想,提出一种新的模糊模型的优化学习算法.将模糊推理规则转化为模糊RBF网络模型,用模糊C均值(FCM)聚类算法和分区效验熵得到模型结构,用混沌变换序列寻优得到优化的中心初值群,用FCM获得最优聚类中心,最后获得模糊神经网络模型.将该方法应用于转炉终点磷含量预报模型,取得了较好的结果.
Based on the idea of chaos optimization, an optimization algorithm for the fuzzy model is presented. The fuzzy model can be represented as a fuzzy RBF neutral network model. The structure of the model is determined using the FCM algorithm and the clustering validity criteria. The initial parameter of clustering centers is obtained using synthetical chaos series and is further optimized using the FCM algorithm. The proposed approach is used successfully for the prediction of end phosphorus content in converter.
出处
《控制与决策》
EI
CSCD
北大核心
2005年第3期261-265,共5页
Control and Decision
基金
国家863计划基金项目(2001AA11040)
湖南省自然科学基金项目(01JJY3029)
湖南省教育厅基金项目(04G718).
