Parameter estimation of dynamic GMDH neural networks with the bounded-error technique

This paper presents a new identification method based on ANNs (Artificial Neural Networks). In particular, a GMDH (Group Method of Data Handling) type neural network whose neurons have hyperbolic tangent activation junctions is considered. For such a network type. a new approach based on a bounded-error set estimation technique is employed to estimate the parameters of the ANN. The final part of this work contains an illustrative example regarding modeling the juice temperature at the outlet of an evaporator at the Lublin Sugar Factory S.A.