基于免疫原理的RBF网络Carrousel氧化沟系统预报  

Immune principle for RBF network and its application on Carrousel oxidation ditch system prediction

作　　者：徐常祝[1] 邓伟[1] 

机构地区：[1]苏州大学计算机科学与技术学院,江苏苏州215006

出　　处：《计算机工程与应用》2009年第26期240-242,共3页Computer Engineering and Applications

基　　金：国家自然科学基金No.60572074~~

摘　　要：为提高Carrousel氧化沟系统水质模拟的精度,利用免疫算法结合RBF神经网络进行建模,并加入k-means聚类对样本进行预处理。以某污水处理中心两年生产数据进行实验,测得出水TN的预报误差0.1956,出水TP的预报误差0.1456,较仅用RBF网络均有很大提高,证明该方法可以应用于Carrousel氧化沟系统的在线实时预测。To improve stimulating precision of water quality for Carrousel oxidation ditch system,immune arithmetic and RBF neural network are combined together in this model.Furthermore,using k-means algorithm as pretreatment step.In the end,doing experiment with a waste water treatment center＇s data,the forecasting error of TN is 0.195 6,and the forecasting error of TP is 0.145 6,the precision improves much compared with RBF network.It proves well that this method can be used in online real time forecasting of Carrousel oxidation ditch system.

关 键 词：免疫算法 RBF神经网络 K-MEANS算法 氧化沟系统 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]