Advanced Signal Processing Techniques for Damage Detection in Reinforced Concrete Beams

Abstract

Structural health monitoring (SHM) using non-destructive testing (NDT) generally involves measurement of shift in natural frequency of the monitored structure. Vibration sensors play a crucial role in such SHM systems and the present-day SHM systems use commercially available off-the-shelf (COTS) micro-electro mechanical system (MEMS) accelerometers. Reduction in natural frequency indicates the degradation of stiffness and consequential damage of the concrete structures. This paper presents the damage detection of reinforced concrete beams using COT MEMS accelerometer. For this work, four reinforced concrete beams of size 125 × 250 × 3200 mm were cast. Two of them were considered as control and two as damaged beams. Damage was inducted to the beams after 28 days of curing using accelerated corrosion set up. All the four beams were subjected to both static and dynamic tests. The maximum load carrying capacity was determined using the static test, and the signals from vibration test of the beams were taken using COTS MEMS accelerometer. The instrumentation of dynamic test includes data acquisition system with signal conditioning circuit, DAQ measurement hardware and a computer with programmable software and multi storage oscilloscope. The vibration signals acquired from the dynamic test is then analysed using signal processing techniques like fast Fourier transform (FFT) and Morlet wavelet transforms and are compared. From the study, it is concluded that the time–frequency contour map helps to identify the damage in a more effective way than that of FFT analysis.