Variable Sampling Time Intelligent Control System For Driving the Hearth of an Industrial Furnace

Abstract

In this paper a PID controller and an adaptive PID controller are designed and implemented in order to control the speed of the induction motors used to operate the rotary hearth furnace with constant load. The PID controller is further discretized and implemented with different values for the sampling time, leading to the online adaptation of the system which is ought to improve the obtained performances of the process. It is possible to implement the adaptive control strategy by means of artificial intelligence techniques which relay on the usage of neural networks – the neural networks are responsible with the generation of the coefficients of the PID controller, which are considered functions that depend on the sampling time. Even if the behavior of the induction motor is strongly nonlinear, the proposed control strategy can follow with high accuracy different setpoint speed values, fact proven through simulations.