基于模糊神经网络PID控制的污水处理应用研究  

Application Research of Sewage Treatment Based on Fuzzy Neural Network PID Control

作　　者：张秀玲[1] 郑翠翠[1] 贾春玉[1] 

机构地区：[1]燕山大学电气工程学院工业计算机控制工程河北省重点实验室,河北秦皇岛066004

出　　处：《化工自动化及仪表》2010年第2期11-13,18,共4页Control and Instruments in Chemical Industry

基　　金：国家自然科学基金资助项目(50675186)

摘　　要：针对活性污泥污水处理系统具有复杂的非线性和时变性,传统的控制方法存在着精度不高,自适应能力差等缺点,提出一种模糊神经网络PID控制方法,将模糊神经网络与PID相结合,既发挥了PID控制的优势,又增加了模糊神经网络自学习和处理定量数据的能力,并且其中采用了动态递归神经网络对污水处理系统进行模型辨识。该控制方法能够快速、有效地使曝气池中溶解氧浓度达到期望值,并且具有较好的控制效果与控制精度。仿真结果验证了该控制方法的有效性和正确性。Aimed at the shortcomings of activated sludge wastewater treatment system with complex non-linear and time-varying, the precision of traditional control method was low and the adaptive capacity was poor, a fuzzy neural network PID control method was put forward. It gave full play to the advantage of PID control and increased the abilities of self-learning and deal with quantitative data by combining fuzzy neural network with PID. A dynamic recurrent neural network was adopted for sewage treatment system model identification. The control method can quickly and effectively make the concentration of dissolved oxygen in aeration tank to meet the expectation value, and has good control performance and control precision. The simulation results verify the effectiveness and correctness of the control method.

关 键 词：模糊神经网络 PID控制 动态递归神经网络辨识 活性污泥法 溶解氧浓度 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]