Application of Neural Networks and PID in on Line of Real-Time Temperature Control System
The Heat Process Trainer PT326 (Feedback UK) model is obtained by using two different techniques one by the Ziegler-Nichols approximating method and the other with the system identification method. The Data acquisition card EC641 I/O PT326 is developed and tested. it can be used as an interface between PC and any similar controlled system. The “velocity” form of the PID controller algorithm is implemented in real time where this algorithm is an ideal of solving the bump less transfer mechanism between manual and automatic control operation and the implementation of anti-reset windup algorithm. The method of back propagation neural network algorithm online gives very good results with neither steady state error nor overshoots.
S. A. Elfandi et al., "Application of Neural Networks and PID in on Line of Real-Time Temperature Control System", Applied Mechanics and Materials, Vols. 110-116, pp. 5009-5014, 2012