Monitoring and flaw detection during wire-based directed energy deposition using in-situ acoustic sensing and wavelet graph signal analysis

•The wire-based directed energy deposition process was monitored using an in-situ acoustic sensor installed near the electric arc.•Onset of flaw formation, such as porosity, variation in line width, and spatter are not readily observable from the raw acoustic sensor data.•Wavelet integrated graph theory approach is used to analyze the data and extract a single feature, called the Fiedler number.•Using the Fiedler number to monitor the process, previously undetected flaws were captured with a false alarm rate less than 2%.