Damage detection of in service timber poles using Hilbert-Huang transform

Abstract

Timber utility poles represent a significant part of Australia's infrastructure for electricity distribution and telecommunication networks. Annual replacement of timber poles are undertaken to avoid pole failures and approximately 40–50 million dollars are spent annually on maintenance and asset management. However, currently practiced inspection techniques for timber poles are mostly subjective in nature and have a lack of reliability, hence, the condition-assessment results provide great difficulties for the asset owners to plan the renewal works. Commonly used wave-propagation based non-destructive testing (NDT) techniques for piles and other concrete structures were adopted and modified to assess the condition of timber poles. The effectiveness of the developed method to detect defects relies on the propagation and reflection of generated shear waves by a transverse impact. However, the orthotropic material properties of timber, the effect of soil embedment, the availability of natural imperfections in timber, and a highly dispersive nature of shear wave propagation make the analysis complex. Therefore, advanced signal-processing techniques are required to identify the damage, the damage location and to accurately assess the damage severity. This paper focuses on damage detection of timber poles using Hilbert-Huang transform as an effective time-frequency analysis technique. A numerical model for an in-situ pole system is developed to simulate the shear wave propagation and to evaluate the capability of the proposed signal-processing technique for an accurate detection of damages. The developed numerical model is validated with the experimental data from the pole yard test and with the data from field-testing. The validated model is then used to simulate different defect profiles, locations and multiple damages to further evaluate the effectiveness of the proposed damage detection technique.