【作者】 徐文； 李宏光；

【Author】 XU Wen,LI Hong-guang (College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China)

【机构】 北京化工大学 信息科学与技术学院；

【摘要】 针对控制图模式识别,提出将控制图判别准则转换为符号化模糊规则,指导模糊神经网络的建模及初始化过程。建立一类融合先验知识的模糊神经网络,给出网络结构设计方法和网络学习算法,对于青霉素发酵过程进行实例研究。研究结果表明:所提出方法的模糊神经网络比传统方法的BP神经网络具有更简单的网络结构、更合理的参数初值,能够获得更准确的控制图模式识别率,具有广阔的应用前景。更多还原

【Abstract】 In regard to control chart pattern recognition,a novel method to convert control chart discriminant criterions is proposed to symbolized fuzzy rules,which help guide to model and initialize a specific fuzzy neural network.Therein,a fuzzy neural network able to combine prior knowledge is explicitly introduced,along with approaches to structure design and training algorithms.Case studies are carried out on a simulation platform concerning penicillin fermentation,leading to satisfying results.The results show that the proposed fuzzy neural networks enjoy simpler structures and more adaptable initial parameters compared to BP networks,which contributes to recognizing control chart patterns accurately as well as expecting more promising applications.更多还原