Statistical methodologies for removing the operational effects from the dynamic responses of a high‐rise telecommunications tower

This paper describes statistical methodologies for removing the influence of operational effects from dynamic responses a telecommunications tower. The characterization structure, over period 3 months, was based on continuous monitoring system that included accelerometers, anemometers and meteorological station. analysis results allowed identifying significant number critical events, which response under wind action is significantly amplified, as well sporadic associated with high peak acceleration values, due to related movement lift, technical staff, equipment. automatic identification extreme required prior removal records using two distinct methodologies, one principal component (PCA) other crest factor (CF) autoregressive models (AR). Both showed efficiency robustness in eliminating peaks effects; however, methodology CF AR proved be computationally more efficient resulted smaller false-positive occurrences events. developed potential integrated Structural Health Monitoring (SHM) systems assess structural safety serviceability towers.