|
NEURAL OUTPUT REGULATION FOR A SOLAR POWER PLANT
J. Henriques, P. Gil and A. Dourado
CISUC - Centro de Informática e Sistemas da Universidade de Coimbra,
Dep. Engenharia Informática, Pólo II da Universidade, Pinhal de Marrocos, 3030 Coimbra, Portugal
e-mail: {jh,pgil,dourado}@dei.uc.pt, Phone: +351 239 79 00 00 Fax:+ 351 239 701 266
In this paper the modelling capabilities of a recurrent neural network and the effectiveness and stability of the output regulation control theory are combined. The control structure consists in a neural based indirect adaptive control scheme, being the main goal to provide a viable practical control strategy suitable for real-time implementations. This control scheme was applied to the distributed solar collector field at Plataforma Solar de Almería, Spain. Experimental results obtained at the solar power plant are presented showing the effectiveness of the proposed approach.
Keywords: Recurrent neural networks, output regulation, non-linear control, on-line learning, solar power plants
Session slot T-Th-M03: Advances in Adaptive Control/Area code 3b : Adaptive Control and Tuning
 | 