A Neural Network-Based Control System Using PID Controller To Control the Deaerator

Abstract

At a steam power station, the pressure and water level in a deaerator are interconnected. Using a common control method based on a traditional PID controller, it is quite difficult to obtain a high degree of pressure and level control in the deaerator. This article proposes a control method based on neural networks that simulate a single PID controller loop. The PID controller has simple and necessary characteristics, as well as high reliability and stability in operation. The neural network, in turn, has the ability to self-learn and to control non-linear processes. Using the proposed method, you can take advantage of both the PID controller and the neural network at the same time. This will significantly reduce overshoot and reduce the required time for transient processes to quickly achieve balance in the control system. Also, the neural network will provide greater stability and a higher response rate when controlling pressure and water level in the deaerator.