Premium
Generalised formulation of composite filters and their application to earthquake engineering test systems
Author(s) -
Stoten David Paul
Publication year - 2017
Publication title -
earthquake engineering and structural dynamics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.218
H-Index - 127
eISSN - 1096-9845
pISSN - 0098-8847
DOI - 10.1002/eqe.2921
Subject(s) - earthquake shaking table , acceleration , earthquake engineering , engineering , transfer function , state space , signal (programming language) , test data , displacement (psychology) , control theory (sociology) , noise (video) , control engineering , computer science , control (management) , structural engineering , mathematics , psychology , statistics , physics , electrical engineering , software engineering , classical mechanics , artificial intelligence , image (mathematics) , psychotherapist , programming language
Summary This paper addresses the problem of generating unmeasured kinetic data—and/or providing improvements in existing data—for the enhancement of performance characteristics of earthquake engineering test systems, such as shaking tables, reaction walls and other custom‐made test rigs. The approach relies upon the use of composite filters (CF), a method of data fusion that was originally conceived via transfer function formulation. The current work generalises the CF concept and extends its formulation into the state‐space domain, thereby providing a wider basis for application to test systems and their controllers, including those of a multivariable (coupled, multi‐axis) nature. Comparative simulation studies of shaking table control are presented that demonstrate the design techniques for state‐space CF and also their effectiveness for signal synthesis, noise suppression and performance improvement. Specific examples include the use of CF for displacement demand signal generation, velocity feedback generation and acceleration control. In each case, the essential principles behind CF—output signals with zero bias and zero drift—are consistently upheld. Copyright © 2017 John Wiley & Sons, Ltd.