Development of a low-power embedded solution for the real-time detection of structural cracks through acoustic emission

This thesis discusses the development of a low-power embedded system capable of performing real-time structural crack detection with acoustic emissions. Firstly, a panoramic analysis of structural health monitoring and acoustic emissions is given, highlighting the most important notions to understand the field. Alongside this study, the crucial and necessary civil engineering, and electronic concepts and components employed in this work are introduced. Building on this foundation the thesis embellishes the most transformative and decisive proceedings in the structural health monitoring field to understand the current state of the art, the most interesting areas of research, and future directions. Thereafter, a detailed description of the embedded solution's evolution is given, showcasing the hardware development's caveats, and the solutions for sampling the acoustic emission signal. During this phase, an analog amplifier circuit is designed to condition the raw signal for easier processing by the selected embedded microcontroller. Afterward, it details the firmware development that is capable of capturing the sensors' acoustic emissions and detecting the structural cracks. The constructed firmware transforms the analog signal into the digital domain, detects structural cracks, and transmits the obtained data through a serial connection to a Python client. Then, the Python script for receiving the information from the embedded platform and visualizing the data is introduced and examined. After the detailed presentation of the methodology, the proof of concepts and the results obtained with the system are presented. The developed embedded solution detects structural cracks generated on a marble block with the industry-standard pencil lead break test. It successfully amplifies, processes, transmits, and visualizes four concurrent acoustic emission signals from 4 sensors in real-time. A performance analysis is given with overall system accuracy and potential real-life applications. Finally, a conclusion is drawn and future development ideas are debated.