Neural networks applied to a multivariable nonlinear control strategies

Abstract

Listen

Translate article

Abstract The Artificial Neural Networks (ANN) are computational tools with great number of applications in modeling and process control techniques. The neural networks can learn sufficiently accurate models and give good nonlinear control when model equations are unknown or only partial state information are available. Neural network approach allows taking into account non-linearities of the process as well as variable interactions. The proposed work has as objective to explore the use of the neural networks in multivariable control strategies in which ANN are used as dynamic models in the generation of predictions, as well as in the definition of adaptive control strategies based on neural networks.