基于神经网络的污水处理鼓风机曝气控制方法  

A neural network-based aeration control method for sewage treatment blower

作　　者：肖梅 XIAO Mei(Department of Public Utilities and Road and Bridge Engineering,Shanghai Communications Polytechnic College,Shanghai,200030,China)

机构地区：[1]上海交通职业技术学院公用事业和路桥工程学院,上海200030

出　　处：《自动化与仪器仪表》2025年第2期116-120,共5页Automation & Instrumentation

摘　　要：针对污水处理鼓风机曝气控制的时滞性,影响污水处理鼓风机曝气控制效果差,提出基于神经网络的污水处理鼓风机曝气控制方法。在分析污水处理鼓风机曝气控制时滞问题后,设计基于神经网络的曝气优化控制模型,该模型主要使用神经网络,根据实时溶解氧浓度数据,结合PID控制器,调整控制器的控制参数,以此驱动PID控制器调节鼓风机变频器输出频率,完成曝气控制,解决时滞问题。模型使用改进粒子群优化算法与BP算法混合训练,更好地适应时滞变化并优化曝气控制策略。该方法使用后,污水处理鼓风机曝气控制的溶解氧浓度处于期望区间,且吨水电耗量减少、水质得到有效优化。Aiming at the delay of aeration control of sewage treatment blower,which affects the poor aeration control effect of sew-age treatment blower,a method of aeration control of sewage treatment blower based on neural network is proposed.After analyzing the delay of aeration control of sewage treatment blower,an aeration optimization control model based on neural network is designed.The model mainly uses neural network to adjust the control parameters of the controller according to the real-time dissolved oxygen concentration data and PID controller,so as to drive the PID controller to adjust the output frequency of blower inverter,complete aeration control and solve the delay problem.The model is trained with improved particle swarm optimization algorithm and BP algo-rithm to better adapt to the change of time delay and optimize the aeration control strategy.Experimental verification:After using this method,the dissolved oxygen concentration controlled by the sewage treatment blower aeration is within the expected range,and the water consumption per ton is reduced,and the water quality is effectively optimized.

关 键 词：神经网络 PID控制器 污水处理 鼓风机 曝气控制 混合训练 

分 类 号：TU992[建筑科学—市政工程] TP273[自动化与计算机技术—检测技术与自动化装置]