模糊PID神经网络算法在污水处理中的应用

溶解氧控制是污水处理厂控制的关键,针对当前溶解氧控制效率低下的问题,提出了一种具有自学习和自适应功能的模糊PID神经网络算法。首先建立PID模糊控制规则表;然后采用神经网络对模糊控制规则表进行不断的训练,得到最佳的模糊控制规则表;最后利用Matlab仿真和工程模拟调试进行验证。结果表明,在大滞后非线性的曝气控制系统中,模糊PID神经网络算法能在最短的时间内使溶解氧达到所期望的值,且整个控制过程具有良好的动态性能和稳态性能。

Application of Fuzzy PID Neural Network in Wastewater Treatment System

Controlling the dissolved oxygen content is crucial in sewage treatment plant. In order to solve the low efficiency of controlling the dissolved oxygen, the fuzzy PID neural network algorithm with self learning and adaptive function was put forward. Firstly, the fuzzy control rule table of PID was established. Then the fuzzy control rules was trained by neural network until the best fuzzy control rules was obtained. Finally, all the techniques were verified by simulation using Matlab and experimental equipment. The results show that in a aeration control system, the fuzzy PID neural network algorithm can make the dissolved oxygen reach the desired value in the shortest time, and the whole control process has good dynamic performance and steady state performance.