基于模糊神经网络的自适应精确曝气系统设计  

Adaptive aeration control scheme design based on fuzzy neural network

作　　者：孙骞[1,2] 许睿[2] 龚琼[2] 

机构地区：[1]桂林市环境保护局,广西桂林541002 [2]桂林电子科技大学生命与环境科学学院,广西桂林541004

出　　处：《桂林电子科技大学学报》2011年第5期382-385,共4页Journal of Guilin University of Electronic Technology

基　　金：国家科技支撑计划(2011BAK12B04);环保部公益性专项(200909060);中国博士后基金(20110491727);广西科学研究和技术开发计划(GKG1140002-2-4)

摘　　要：基于模糊神经网络理论,将自回归各态历经(ARX)模型与模糊神经网络模型(FNNM)相结合,设计一套具有自适应能力的前馈-反馈精确曝气控制系统,解决目前污水处理曝气过程中冗余曝气能源浪费、溶解氧波动大的问题。以污水厂处理PID控制生物池数据作为训练样本,以模糊神经网络自适应精确曝气模型仿真与传统PID控制比较,发现该系统稳定时间较短、溶解氧浓度波动幅度更小。Based on fuzzy neural network theory, auto regressive ergodic （ARX） model and fuzzy neural network model （FNNM） were combined to design an adaptive feed forward-feedback precise aeration control system for re- ducing redundant energy waste caused by aeration and dissolved oxygen fluctuations in the current wastewater treat- ment aeration process. Wastewater treatment plant biological pool data as training samples, the adaptive fuzzy neu- ral network model accurately aeration was compared with the traditional PID control, the results show that the set- tling time is faster, and the fluctuations of dissolved oxygen concentration is smaller in the system.

关 键 词：模糊神经网络模型 曝气 时滞 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]