一种基于PSO优化带核函数FKM的高炉炉温辨识方法

An Identification Method of Blast-Furnace Temperature Based on FKM with Kernel Function Optimized by PSO

提出了一种高炉炉温分类辨识的方法PSO-K-FKM,该方法采用基于核函数改进的模糊K均值聚类,利用粒子群对聚类中心寻优.将该方法应用于攀钢3#高炉炉温分类判定,结果表明:该方法基本上能够正确划分高炉炉温状况.与BP方法、GA方法相比,该方法具有识别率高、识别速度快、不易陷入局部最优等优点.

Particle swarm optimization （PSO） algorithm is adopted to optimize the center of fuzz K-means （FKM） with kernel function, which is used to classify the temperature of blast furnace. This PSO-FKM method takes advantage of colony to find new avenue for the solution of complex problems and profits from the character of speediness and efficiency of K-means. Experiments show that it has higher cognition, compared to other methods such as BP and SW-GA. This method can obtain a good result in practice for 3 # blast furnace of Panzhihua Iron and Steel Industry Co. Ltd.