Experimental modeling of a pneumatic system


Zorlu A., Ozsoy C., Kuzucu A.

9th IEEE International Conference on Emerging Technologies and Factory Automation, Lisbon, Portugal, 16 - 19 September 2003, pp.453-461 identifier

  • Publication Type: Conference Paper / Full Text
  • City: Lisbon
  • Country: Portugal
  • Page Numbers: pp.453-461

Abstract

To know system model, structure and real parameters very well will increase the success of the control to be applied on it. This paper presents an experimental modelling study on a pneumatic system. PRBS signals are applied to the servovalves and system responses as position and chamber pressures are recorded. Later these data are digitally processed and input-output data for identification models are created. Pneumatic system model consists of a combination of two pressure models which each of them as two inputs and one output, defining pressure changes in the chambers and a velocity model which has one input-one output, defining piston motion. Parameters of the pressure models and the velocity model that are installed considering the mathematical equations of the system are identified by using Adaptech Midsys identification software. So, discrete-time models are obtained. Later they are converted into continuous-time models. During the identification, extended least squares method, parameter adaptation algorithm with decreasing gain and canonical parametrization of Guidorzi have been applied. As a result, parameters of the pneumatic system such as friction coefficients, moving mass, valve coefficients etc. are identifiable.