【作者】 王妍； 李宏光；

【Author】 WANG Yan,LI Hong-guang (School 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】 针对青霉素发酵间歇过程,探讨基于赋时Petri网和RBF神经网络相结合的过程特征状态监督方法。通过过程分析,描述若干个过程的特征状态,建立2个RBF神经网络,分别对主特征状态和辅助特征状态进行提取。采用赋时Petri网建立特征状态的演化模型,从而实现对间歇过程的实时智能监督。在青霉素发酵仿真实验对象上进行验证,结果表明了技术方法的有效性。更多还原

【Abstract】 In regard to penicillin fermentation batch processes,a characteristic state based supervision approach which combined RBF neural network and timed Petri net techniques was presented.Initially,a couple of characteristic states were specified through the process analysis.Two RBF neural network systems were accordingly established to extract the main characteristic states and the auxiliary ones,respectively.After that,a timed Petri net based characteristic state evolutionary model was built,which enabled real-time intelligent monitoring of the batch processes.Experimental studies were carried out on a penicillin fermentation simulation platform,demonstrating the effectiveness of the proposed approaches.更多还原