Numerical modeling of the hydraulic GEROLER motor using the artificial neural network


  • Goran Gregov University of Rijeka, Faculty of Engineering, Vukovarska 58, 51000 Rijeka, Croatia


This paper deals with the numerical modeling of the hydraulic GEROLER motor. GEROLER motors are known for their cost effectiveness and balance between simplicity, robustness, compactness, versatility, and noise. The analysis was carried out using a black-box approach taking into account the nonlinear dynamic behavior of the GEROLER motor. The artificial neural networks method was used to define the black-box model. Two models of neural networks were used: the first one comprising multi-layer feed-forward neural networks and the second one the NARX dynamic neural networks. The results obtained by the numerical simulations of the GEROLER motor model were compared with the experimental measurements performed on a laboratory hydraulic system. The derived model has provided results that allow a high degree of a generalized approach to the motor design.