Bolted joints are among the most common building blocks used across different types of structures, and are often the key components that sew all other structural parts together. Monitoring and assessment of looseness in bolted structures is one of the most attractive topics in mechanical, aerospace, and civil engineering. This paper presents a new percussion-based non-destructive approach to determine the health condition of bolted joints with the help of machine learning. The proposed method is very similar to the percussive diagnostic techniques used in clinical examinations to diagnose the health of patients. Due to the different interfacial properties among the bolts, nuts and the host structure, bolted joints can generate unique sounds when it is excited by impacts, such as from tapping. Power spectrum density, as a signal feature, was used to recognize and classify recorded tapping data. A machine learning model using the decision tree method was employed to identify the bolt looseness level. Experiments demonstrated that the newly proposed method for bolt looseness detection is very easy to implement by ‘listening to tapping’ and the monitoring accuracy is very high. With the rapid in robotics, the proposed approach has great potential to be implemented with intimately weaving robotics and machine learning to produce a cyber-physical system that can automatically inspect and determine the health of a structure.

In this paper, an innovative method for rapid detection and location determination of pipeline leakage utilizing lead zirconate titanate (PZT) sensors is proposed. The negative pressure wave (NPW) is a stress wave generated by leakage in the pipeline, and propagates along the pipeline from the leakage point to both ends. Thus the NPW is associated with hoop strain variation along the pipe wall. PZT sensors mounted on the pipeline were used to measure the strain variation and allowed accurate (within 2% error) and repeatable location (within 4% variance) of five manually controlled leakage points. Experimental results have verified the effectiveness and the location accuracy for leakage in a 55 meter long model pipeline.

Reports indicated that impact events accounted for 47% of offshore pipeline failures, which calls for impact detection and localization for subsea pipelines. In this paper, an innovative method for rapid localization of impacts on underwater pipelines utilizing a novel determination technique for both arrival-time and group velocity (ATGV) of ultrasonic guided waves with lead zirconate titanate (PZT) transducers is described. PZT transducers mounted on the outer surface of a model pipeline were utilized to measure ultrasonic guided waves generated by impact events. Based on the signals from PZT sensors, the ATGV technique integrates wavelet decomposition, Hilbert transform and statistical analysis to pinpoint the arrival-time of the designated ultrasonic guided waves with a specific group velocity. Experimental results have verified the effectiveness and the localization accuracy for eight impact points along a model underwater pipeline. All estimations errors were small and were comparable with the wavelength of the designated ultrasonic guided waves. Furthermore, the method is robust against the low frequency structural vibration introduced by other external forces.

In this paper, an innovative method for real-time estimating impact location on concrete structures utilizing lead zirconate titanate (PZT) sensors is proposed. P-waves are a kind of stress waves excited by the impact events on concrete structures, and propagate in concrete structures with the information of the impact location. PZT sensors embedded in the concrete column were utilized to measure the P-waves and locate the impact locations with acceptable accuracy and precision. Experimental results have verified the effectiveness and the location accuracy for 15 simulated impact points on a model concrete column. Compared to existing methods, which rely on low energy acoustic emissions, the proposed method can detect high energy impulse events, typically resulting from impacts. Furthermore, the method is resistant against the effects of low frequency whole-structure vibrations introduced by the impact force. All of the above can be accomplished through a sparse array of sensors, which makes the method minimally intrusive.

No abstract