Development of modified LQG controller for mitigation of seismic vibrations using swarm intelligence

Gaurav Kumar, Roshan Kumar, Ashok Kumar, Brij Mohan Singh

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)


A method is presented to design and tune the modified linear quadratic Gaussian (LQG) controller to obtain increased efficiency during an earthquake. It utilises swarm intelligence to tune the parameters of LQG based on quasi resonance between the natural frequencies of the structure in first two modes and the predominant frequencies of the seismic signal. The modified controller thus developed minimises the energy of structure by altering its parameters online. For testing of this modified controller, a benchmark prototype structure is numerically tested under different seismic signatures recorded in near/far fault sites in the different soil conditions. A parametric study comparing the efficiencies of modified LQG, and other contemporary controllers is presented. It is observed for El-Centro earthquake that the modified controller achieved reductions of 22%, 33% and 27% in relative displacement, inter-storey drift, and absolute acceleration respectively as compared to the conventional LQG controller. Similar results are observed for Gebze and Chi-Chi earthquakes. The modified controller is also evaluated in a situation where power vanishes at the peak of the seismic excitation. Based on the results and discussion, the performance of the proposed controller is observed to be superior among all controllers considered in this study.

Original languageEnglish
Pages (from-to)19-42
Number of pages24
JournalInternational Journal of Automation and Control
Issue number1
Publication statusPublished - 2022

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Hardware and Architecture
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Development of modified LQG controller for mitigation of seismic vibrations using swarm intelligence'. Together they form a unique fingerprint.

Cite this