Neural Network in Scheduling Linear Controllers With Application to a Solar Power Plant

This work presents a hybrid scheme combining the potentialities of neural networks for approximation purposes with the well-know theory and widespread industrial application of PID techniques. The neural network is trained based on measured data from the plant providing a way of scheduling between a set of PID controllers, a priori tuned in different operating points by means of Takahashi rules. This neural network control strategy is in practice applied to the control of a distributed collector field of a solar power plant. Experimental results collected at Plataforma Solar de Almeria (Spain), show the effectiveness of the proposed approach.