CMAC based controller for hydro-mechanical systems

Abstract

A cerebellar model articulation controller (CMAC) neural network based scheme for controlling a non-linear mechanical system that incorporates hydraulic actuator dynamics is proposed in this paper. This control scheme has been shown to perform well in the presence of non-linearities due to fluid flow, oil compressibility. leakages and static friction. Further, the algorithms developed in this paper have been shown to effectively track different trajectories and reject disturbances while compensating uncertain dynamics present in the hydro-mechanical system. This control scheme also shows that it can cope with complexities associated with a highly non-linear hydraulic system, which is more desirable than conventional control schemes that are typically designed using linearised models.

Velocity control of an electro hydraulic servomotor by neural networks

“…If the state variables are denoted by: 1 x -hydro motor angular velocity 2 x -load pressure differential 3 x -valve displacement Then the model of EHSS is given by: The control objective is stabilization of any chosen operating point of the system. It is readily shown that equilibrium points of system are given by:…”