一种基于模糊准则的发酵过程残糖质量浓度的预测方法  

Predictive Algorithm for Residual Sugar Concentration in Fermentation Process Based on Fuzzy Rule

作　　者：刘辉 赵忠盖 刘飞 LIU Hui;ZHAO Zhonggai;LIU Fei(Key Laboratory of Advanced Process Control for Light Industry,Ministry of Education,Jiangnan University,Wuxi 214122,China)

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

出　　处：《食品与生物技术学报》2018年第9期931-938,共8页Journal of Food Science and Biotechnology

基　　金：国家自然科学基金项目(61573169);江苏省六大人才高峰项目(2014-ZBZZ-010)

摘　　要：发酵过程残糖质量浓度的在线检测通常具有较大滞后,极大影响了残糖质量浓度的控制。发酵工艺人员根据菌体生长耗糖的惯性特征,设计了一种残糖质量浓度的经验估算算法,但是在残糖质量浓度变化较大时,估算效果不理想。作者首先分析工艺人员的经验估算算法的缺点,结合反馈校正思想,提出了经验估算改进(即有补偿)算法,接着针对工艺人员的残糖质量浓度估算模型的不准确性,结合模糊智能技术,进行模糊模拟,分别研究模糊预测无补偿、模糊预测有补偿情况下,残糖质量浓度的预测情况,最后通过实验验证了模糊残糖质量浓度估算模型以及补偿算法的准确性和有效性。In the fermentation process,it is difficult to implement a good online control of residual sugar concentration with delayed measurements.Due to the inherit characteristics of biomass growth,the field operator can achieve an effective estimation on the residual sugar concentration according to operating experience or knowledge.However,it is ineffective in the large fluctuation of the residual sugar concentration.Based on the operating experience or knowledge,this paper introduces a feedback correction idea to compensate the experience estimation algorithm.Moreover,considering the inaccuracy model of the residual sugar concentration,this paper proposes a fuzzy predictive estimation algorithm,a fuzzy predictive estimation algorithm with compensation combining fuzzy intelligent technology.Afterwards,the residual sugar concentration is forecast by these two predictive estimation algorithms.The accuracy and effectiveness of the fuzzy model and compensation algorithm are verified by a real experiment.

关 键 词：发酵过程 残糖浓度 滞后 模糊 预测 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]