Hidden Markov models for pipeline damage detection using piezoelectric transducers

Oil and gas pipeline leakages lead to not only enormous economic loss but also environmental disasters. How to detect the pipeline damages including leakages and cracks has attracted much research attention. One of the promising leakage detection method is to use lead zirconate titanate (PZT) transducers to detect the negative pressure wave when leakage occurs. PZT transducers can generate and detect guided waves for crack detection also. However, the negative pressure waves or guided stress waves may not be easily detected with environmental interference, e.g., the oil and gas pipelines in an offshore environment. In this paper, a Gaussian mixture model-hidden Markov model (GMM-HMM) method is proposed to process PZT transducers’ outputs for detecting the pipeline leakage and crack depth in changing environment and time-varying operational conditions. Leakages in different sections or crack depths are considered as different states in hidden Markov models (HMMs). One time-domain damage index and one frequency domain damage index are extracted from signals collected from PZT transducers, then extracted indices are formed as observation emissions in the HMM. The observation probability distribution matrix in HMM is initialized by a Gaussian mixture model (GMM) to address signal uncertainties. After the HMM parameter initialization, an iterative training process through the Baum–Welch algorithm is applied to get the optimized parameters of the GMM-HMM. Leakage location or crack depth is decided by the maximum posterior probability from the trained model. Two different experimental settings and results show that the GMM-HMM method can recognize the crack depth and leakage of pipeline such as whether there is a leakage, where the leakage is.