Wavelet Power and Shannon Entropy Applied to Acoustic Emission Signals for Corrosion Detection and Evaluation of Reinforced Concrete
Abstract
Acoustic emission (AE) signals detected from corrosion test on a steel reinforced concrete beam subjected to the coupling effects of corrosive wet-dry cycles and static load are analyzed by power spectral density, wavelet transform, and Shannon entropy. The degradation process of the corroded reinforced concrete beam can be divided into four stages on the basis of the accumulated event number (AEN). Due to the difference of material properties, steel reinforcement and concrete matrix have distinguished AE features. The time-frequency characteristics of AE signals can reflect the microstructural degradation mechanism of steel corrosion and concrete cracking. The corrosion evaluation entails investigating the evolution of the wavelet power mathematically by Shannon entropy. The frequency-entropy clearly exhibits the relative power distribution of AE signal in a certain frequency region. With the accumulation of steel corrosion and concrete deterioration, the increment of the overall entropy integration is considerably apparent. The variation of frequency-entropy curve reveals the corrosion revolution of the reinforced concrete members under static load, which is represented by a transforming from corrosion-induced micro cracking to load-induced localized cracking.
