Discrete PID Controller Tuning Using Piecewise-Linear Neural Network

PID controller (which is an acronym to “proportional, integral and derivative”) is a type of device used for process control. As first practical use of PID controller dates to 1890s (Bennett, 1993), PID controllers are spread widely in various control applications till these days. In process control today, more than 95% of the control loops are PID type (Astrom et al., 1995). PID controllers have experienced many changes in technology, from mechanics and pneumatics to microprocessors and computers. Especially microprocessors have influenced PID controllers applying significantly. They have given possibilities to provide additional features like automatic tuning or continuous adaptation – and continuous adaptation of PID controller via neural model of controlled system (which is considered to be significantly nonlinear) is the aim of this contribution. Artificial Neural Networks have traditionally enjoyed considerable attention in process control applications, especially for their universal approximation abilities (Montague et al., 1994), (Dwarapudi, et al., 2007). In next sections, there is to be explained how to use artificial neural networks with piecewise-linear activation functions in hidden layer in controller design. To be more specific, there is described technique of controlled plant linearization using nonlinear neural model. Obtained linearized model is in a shape of linear difference equation and it can be used for PID controller parameters tuning.