三相生物流化床味精废水处理的优化控制研究  

Research on Optimal Control of MSG Wastewater Treatment Based on Three-Phase Biological Fluidized Bed

作　　者：连晓峰[1] 刘载文[1] 刘萌[1] 王磊[1] 王小艺[1] 

机构地区：[1]北京工商大学计算机与信息工程学院,北京100048

出　　处：《北京工商大学学报（自然科学版）》2011年第2期49-53,共5页Journal of Beijing Technology and Business University:Natural Science Edition

基　　金："十一五"国家科技支撑计划项目(2007BAK36B07);北京工商大学青年教师科研启动基金项目(2009)

摘　　要：针对三相生物流化床味精废水处理过程的闭环控制进行研究,在考虑其大滞后性、复杂非线性等特点的基础上,提出一种基于人工神经网络的非线性模型预测控制方法以实现处理过程的闭环优化控制.该方法首先建立味精废水处理工艺模型,采用非线性模型预测控制方法对模型进行校正,并利用序列二次规划法解决其约束非线性规划问题,最终实现味精废水处理的闭环优化控制.仿真和实验结果表明该方法有效,可实现闭环实时控制,具有广阔的应用前景.The close-loop control in the MSG wastewater treatment system based on three-phase biological fluidized bed was studied in this paper.A nonlinear model predictive control method based on ANN was established considering the characteristics of large time-delays and complex nonlinearity.First the chemical model of MSG wastewater treatment process was built up.Next the model was modified by using the nonlinear model predictive control method.Then the sequential quadratic programming method was employed to tackle the constrained high nonlinear programming problems.Finally the close-loop optimal control system was realized.Simulation experiment results showed that this method was effective and could realize the closed-loop real-time control which would have a broad prospect of application.

关 键 词：味精废水 三相生物流化床 非线性模型预测控制 序列二次规划 人工神经网络 

分 类 号：TS201.6[轻工技术与工程—食品科学] TN391.41[轻工技术与工程—食品科学与工程]