Design and Performance Analysis of Incremental Networked Predictive Control Systems

Zhong-Hua Pang, Donghua Zhou, G-P Liu

Research output: Contribution to journalArticlepeer-review

34 Downloads (Pure)

Abstract

This paper is concerned with the design and performance analysis of networked control systems with network-induced delay, packet disorder, and packet dropout. Based on the incremental form of the plant input-output model and an incremental error feedback control strategy, an incremental networked predictive control (INPC) scheme is proposed to actively compensate for the round-trip time delay resulting from the above communication constraints. The output tracking performance and closed-loop stability of the resulting INPC system are considered for two cases: 1) plant-model match case and 2) plant-model mismatch case. For the former case, the INPC system can achieve the same output tracking performance and closed-loop stability as those of the corresponding local control system. For the latter case, a sufficient condition for the stability of the closed-loop INPC system is derived using the switched system theory. Furthermore, for both cases, the INPC system can achieve a zero steady-state output tracking error for step commands. Finally, both numerical simulations and practical experiments on an Internet-based servo motor system illustrate the effectiveness of the proposed method.

Original languageEnglish
Pages (from-to)1400-1410
Number of pages11
JournalIeee transactions on cybernetics
Volume46
Issue number6
Early online date14 Jul 2015
DOIs
Publication statusPublished - Jun 2016

Keywords

  • Experiment
  • networked control systems (NCSs)
  • performance analysis
  • predictive control
  • round-trip time (RTT) delay
  • stability analysis
  • CONSTRAINED NONLINEAR-SYSTEMS
  • TRACKING CONTROL
  • STABILITY ANALYSIS
  • TIME-DELAY
  • IMPLEMENTATION
  • CONSENSUS
  • DROPOUTS

Fingerprint

Dive into the research topics of 'Design and Performance Analysis of Incremental Networked Predictive Control Systems'. Together they form a unique fingerprint.

Cite this