Optimization of synthesis parameters of adaptive generalized predictive control using genetic algorithms

This paper concentrates on the off-line optimization of adaptive generalized predictive control (AGPC) tuning parameters using genetic algorithms (GAs). The Adaptive generalized predictive control algorithm is detailed and presented. A single input single output (SISO) controlled autoregressive integrated moving average (CARIMA) model is used to predict the future behavior of the system. The online parametric adaptation algorithm (PAA) used in this work is based on recursive least squares (RLS) identification algorithm with fixed forgetting factor. The synthesis parameters (minimal prediction horizon, maximal prediction horizon, control horizon and cost weighting factor) are optimized using genetic algorithms, this technique improves the closed-loop performances. The fitness function to be minimized is a set of closed loop performance metrics (the settling time, the rise time, the overshoot and the variance of the control signal). In order to verify the validity of the proposed strategy, a simulation example of asynchronous motor speed is presented. The obtained results shows the effectiveness of this approach.