Affine Neural Network-Based Predictive Control Applied to a Distributed Solar Collector Field

Paulo José Carrilho de Sousa Gil, Jorge H. Henriques, Alberto J.L. Cardoso, Paulo Roberto Antonacci Carvalho, Antonio Dourado

Research output: Contribution to journalArticlepeer-review

23 Citations (Scopus)


This paper presents experimental results concerning the control of a distributed solar collector field, where the main objective concerns the regulation of the outlet oil temperature by suitably manipulating the oil flow rate. This is achieved by means of a constrained nonlinear adaptive model-based predictive control framework where the control action sequence is obtained by solving an open-loop optimization problem, subject to a set of constraints. The plant dynamics is approximated by an affine state-space neural network, whose complexity is specified in terms of the cardinality of dominant singular values associated with a subspace oblique projection of data-driven Hankel matrices. The neural network is first trained offline and subsequently improved through a recursive updating of its weights and biases, based on a dual unscented Kalman filter. The control scheme is implemented on the Acurex field of the Plataforma Solar de Almería, Spain. Results from these experiments demonstrate the feasibility of the proposed framework, and highlight the ability to cope with time-varying and unmodeled dynamics, under the form of disturbances, and its inherent capability for accommodating actuation faults.
Original languageEnglish
Pages (from-to)585-596
JournalIEEE Transactions on Control Systems Technology
Issue number2
Publication statusPublished - Feb 2014


  • Adaptive control
  • affine state-space neural networks
  • constrained optimization
  • distributed solar collector field (DSC)
  • model-based predictive control (MPC)
  • online training
  • unscented Kalman filter


Dive into the research topics of 'Affine Neural Network-Based Predictive Control Applied to a Distributed Solar Collector Field'. Together they form a unique fingerprint.

Cite this