基于LS-SVM模糊推理的发酵过程流加控制  

Optimized control of fed-batch fermentation process based on fuzzy inference of LS-SVM

作　　者：张相胜 苏青[2] 潘丰 

机构地区：[1]江南大学,轻工过程先进控制教育部重点实验室,江苏无锡214122 [2]日照职业技术学院,山东日照276800

出　　处：《计算机与应用化学》2012年第9期1085-1088,共4页Computers and Applied Chemistry

基　　金：国家高技术研究发展计划(863)资助项目(2009AA05Z203);江苏高校优势学科建设工程资助项目

摘　　要：根据发酵过程中各阶段具有模糊性的特点,采用最小二乘支持向量机(1east square support vector machines,LS-SVM)提取判断发酵阶段的模糊规则,把能够在线测量的物理量如发酵时间、pH、CO_2生成速率和NH_3的添加量作为输入,通过模糊推理机判定发酵过程所处的阶段,然后依据加权平均的方法确定各个阶段葡萄糖的流加策略,实现分段连续流加补料控制。通过实验,验证了该方法的有效性。According to the characteristics of fuzzy during the stage of fermentation process, a fuzzy inference method based on least squares support vector machines is proposed, which can be distilled and simplified the fuzzy rule. Utilizing the on-line measurement of physical quantities such as fermentation time, pH and the amount of NH3, the Fuzzy Inference Engine determine the stage fermentation process, and then the glucose-plus strategy is obtained based on the method of weighted average every time, the stability control of the sub-fed-batch continuous is completed. The effectiveness of the method is proved by practical application.

关 键 词：最小二乘支持向量机 模糊推理 流加控制 发酵过程 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.9[自动化与计算机技术—控制科学与工程]