Modelling and simulation of an air motor system using extended radial basis algorithm

Abstract

This article proposes a new modeling scheme using extended radial basis function (RBF) and adaptive neuro-fuzzy filter for handling nonlinear uncertainties of an air motor servo valve. This model combines the fast model development ability of RBF and the adaptation capability of adaptive neuro-fuzzy inference system (ANFIS) used instead of the well known conventional modeling techniques. The ANFIS structure provided parameter partitioning and better performance under transient response to handle the problem of disturbance attenuation. The pneumatic H-bridge, characterizing a pneumatic servo valve has been devised for speed and direction control of the motor and the system characteristics conveniently divided into three main regions; of low speed (below 390 rev/min), medium speed (390 to 540 rev/min) and high speed (540 to 680 rev/min). The system is highly non-linear in the low speed region and hence the need to use an adaptive intelligent based modeling technique arises. Simulation results has proven that for an air motor system with uncertainty and perturbed noise, the RBF-ANFIS model scheme performed well and out past its conventional counterpart by far.