01-04-2014, 10:46 AM
Time-Delay Estimation
Abstract
This is the first part of a two part paper devoted to introducing Dominant Gain (DG) concept in dynamical systems, (part I) and applying it as a constraint in controlling Time-Delay (TD) systems for improving control performances, (part II). In the first part of the paper, after describing the dynamical characteristics of Quasi-Rational-Distributed System (QRDS), based on the model structure of such systems, the DG concept is proved to show that the various characteristics of QRDS models can be exploited and categorized by the DG concept. For further implying and insisting on the obtained results from the DG concept, parameter identification of simplified models for QRDSs were carried out to show that such systems can represent both the minimum phase as well as non-minimum phase characteristics. Further, it was shown that, contrary to the rational models, in QRDS models the Time-Delay parameter affects the gain plot and it can be identified straightly from the period of oscillations appearing in the frequency response data of such systems.
Introduction
The appearance of periodic resonance in both the amplitude ratio and phase diagrams is an indication of irrational form of the structure of a system model [1].
Generally In classic control, systems are classified as rational or irrational process system models. Distributed parameter process systems or equivalently quasi-rational distributed systems (QRDS) are a class of distributed parameter systems in which hyperbolic partial differential models describe their dynamics [2-4] and process variables vary in space as well as in time. It was claimed that almost all natural and industrial processes were are inherently distributed in nature [9]. Important industrial examples in process industries include Heat Exchangers, Tubular and Packed Bed Reactors, Rotary Drums, as well as the dynamics of the so called particulate systems, like Crystallizers and etc. [1-8].
conclusion
Distributed parameter process systems such as heat exchangers and rotary drums (such as rotary dryers and rotary cement kiln etc.) demonstrate a QRDS (Quasi-Rational-Delay) model structure in their transfer function models. A QRDS model is known to express complicated delay characteristics. This can be expressed according to the number of their LHP and RHP zeros. Basically four distinguishable zeros location characteristics are detectable for a QRDS model structure.