An acceleration-oriented form of simple piecewise linearisation in time series to assess seismically damaged structures

Ryuta Enokida, Koichi Kajiwara

Research output: Contribution to journalArticlepeer-review

Abstract

This study presents an acceleration-oriented form of simple piecewise linearisation in time series (SPLiTS) to assess the condition of a seismically damaged structure using only its measured acceleration. Its original form could estimate the physical parameters of nonlinear structures in the time domain using inversions of the displacement and acceleration, based on its piecewise linearisation. However, its reliance on measured displacement limited its application only to structures in heavily monitored environments, such as laboratories. To enhance its feasibility for structures with fewer sensors or improper displacement measurement cases, an acceleration-oriented form is introduced, which does not require displacement measurements. To maintain the procedure’s simplicity, the new form retains the basic signal processing techniques: integrations of acceleration and a multi-pass moving-average filtering technique, to obtain the displacement and velocity responses used in the inversion. Based on the principle of SPLiTS, which minimises the central-point shift components, the average filtering technique removes the distortion generated during integration. The new form was examined by applying it to E-Defense shake table experiments on a three-storey steel structure, which contains an improper displacement measurement case. Although the original and new forms reasonably estimated the physical parameters in proper measurement cases, only the new form was effective in the improper case. The examinations confirmed the effectiveness of the acceleration-oriented form relying on the basic techniques and its applicability to estimating physical parameters of the seismically damaged structure for its condition assessment.

Original languageEnglish
Article number1129403
JournalFrontiers in Built Environment
Volume9
DOIs
Publication statusPublished - 2023 Jan 23

Keywords

  • central-point shift
  • moving-average filter
  • structural energy absorption
  • structural health monitoring
  • system identification
  • time-domain inversion

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Building and Construction
  • Urban Studies

Fingerprint

Dive into the research topics of 'An acceleration-oriented form of simple piecewise linearisation in time series to assess seismically damaged structures'. Together they form a unique fingerprint.

Cite this