Open AccessAnomaly detection and prediction of high‐tension bolts by using strain of tower shell
Author(s) -
Kikuchi Yuka,
Ishihara Takeshi
Publication year - 2020
Publication title -
wind energy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.743
H-Index - 92
eISSN - 1099-1824
pISSN - 1095-4244
DOI - 10.1002/we.2551
Subject(s) - mahalanobis distance , tower , structural engineering , finite element method , shell (structure) , tension (geology) , engineering , taguchi methods , anomaly detection , anomaly (physics) , mechanical engineering , computer science , materials science , composite material , ultimate tensile strength , physics , artificial intelligence , condensed matter physics
In this study, an anomaly detection and prediction method for high‐tension bolts is proposed by using the measured strain of tower shell. The finite element models (FEMs) for the high‐tension bolts and tower‐top are built and validated with measurement data. An algorithm of anomaly detection and prediction is then developed by applying Mahalanobis‑Taguchi system. The threshold of abnormal condition for high‐tension bolts is proposed, and the formula for the prediction of remaining axial bolt force is derived by using the strain of tower shell obtained from the FEM model. Finally, onsite measurements are performed at Taikoyama wind farm to investigate the relationship between the strain of tower shell and the remaining axial bolt force. It is found that the proposed threshold based on the FEM model succeeds in detecting abnormal bolt by the Mahalanobis‑Taguchi system, and the proposed formula accurately predicts the remaining axial bolt force.