Neural Networks for Optimal Selection of The PID Parameters and Designing Feedforward Controller


Abstract:A neural network-based feedforward controller and self-tuning PID controllerwith optimization algorithm is presented. The scheme of the controller is based on twounknown models that describe the system and optimization algorithm. These models aremodified Elman recurrent neural network and NARMA-L2. The modified Elmanrecurrent neural network (MERNN) model and NARMA-L2 model are learned withtwo stages off-line and on-line, in order to guarantee that the output of the modelaccurately represents the actual output of the system. The aim from the NARMA-L2model is to find the Inverse Feedforward Controller (IFC) which controls the steadystateoutput of the system. The MERNN model after being learned is called theidentifier. The feedback PID self tuning control signal for N-step ahead can becalculated the PID parameters by using the optimization algorithm with the quadraticperformance index which is quadratic in the error between the desired set point and themodel output, as well as quadratic of the control action. The paper explains thealgorithm for a general case, and then a specific application on non-linear dynamicalplant is presented.