Damage stage prediction of reinforced concrete under axial compression based on acoustic emission and a hidden Markov model

This paper addresses the challenges in assessing the health status of reinforced concrete (RC) components in large civil structures by proposing a novel method for identifying damage stages. The proposed approach leverages the cumulative energy slope of acoustic emission (AE) signals and a hidden Markov model (HMM) to predict the damage stages. Three main innovations are highlighted: firstly, to improve the accuracy of damage stage prediction, the integration of 'cumulative energy-time' data from AE into a HMM is introduced. Secondly, to overcome the challenge of prediction failure under varying loading rates, a method for unifying different loading rates is proposed. Thirdly, the proposed method facilitates high-precision damage stage prediction at any given moment without needing complete process damage information. This provides valuable insights for estimating the remaining service life of structures.