Adaptive PID controller design by using adaptive interaction approach theory

Tayfun Gundogdu, Guven Komurgoz

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

A self-tuning algorithm for PID controller based on adaptive interaction approach efficiently used in the Artificial Neural Networks (ANNs) is proposed in this paper. The principle behind the adaptation algorithm is mathematically isometric to the back-propagation algorithm (BPA). By applying Adaptive Interaction (AI), the same adaptation as the well-known BPA can be achieved without the need of a feed-back network. Hereby, by using AI tuning algorithm, the ANN PID controller can be adapted directly without wasting calculation time in order to increase the frequency response of the controller. Speed control of a DC motor under the rapidly changing load condition is simulated to demonstrate the sensitivity of the AI algorithm. PID gains of the ANN controller was tuned directly by using AI tuning algorithm. Simulation results and PID adaptation process have been presented.

Original languageEnglish
Title of host publication2013 3rd International Conference on Electric Power and Energy Conversion Systems, EPECS 2013
DOIs
Publication statusPublished - 2013
Event2013 3rd International Conference on Electric Power and Energy Conversion Systems, EPECS 2013 - Istanbul, Turkey
Duration: 2 Oct 20134 Oct 2013

Publication series

Name2013 3rd International Conference on Electric Power and Energy Conversion Systems, EPECS 2013

Conference

Conference2013 3rd International Conference on Electric Power and Energy Conversion Systems, EPECS 2013
Country/TerritoryTurkey
CityIstanbul
Period2/10/134/10/13

Keywords

  • Adaptive Interaction
  • DC motor control
  • PID controller
  • adaptive neural network
  • self-tuning

Fingerprint

Dive into the research topics of 'Adaptive PID controller design by using adaptive interaction approach theory'. Together they form a unique fingerprint.

Cite this