The Simplest Identification Model of an Asymmetric Hydraulic System

摘要

A single-rod cylinder servo system driving a discontinuous load is a 4th-order nonlinear system. It can be simplified as a time-varying 1st-order system for typical low frequency applications. The time-varying parameters of such a system conceal the discontinuous loading effect, uncertainties and transient behavior of a single-rod cylinder and servovalve. To overcome these difficulties, the parameter identification process requires the characteristics of exponential convergence and unbiased parameter estimation. Here, the parameter identification process incorporates the 1st-order ARMAX model with PEDF algorithm. Experiments indicate quantatively that this identification process is superior and more adaptive than high-order ARMA models with RLS algorithm.