Online Fault Detection for Metal Additive Manufacturing with Data-Driven Time Series Models

Process faults pose a significant challenge for metal additive manufacturing (AM), resulting in structurally compromised parts or a print failure. To increase reliability and confidence in the metal AM process, this work presents a dynamic data-driven application system (DDDAS) detection framework that leverages a data-driven time series model to predict the nominal behavior. In-situ images of the melt pool are used to monitor the quality of the weld and the sequence of melt pool area is used as the time series to be modeled. The deviation between the online, monitored melt pool size and the predicted melt pool size is used for defect detection; a large prediction error indicates the corresponding image deviates from the expected. This distinction can be performed automatically using a statistical threshold metric, which forms the basis of a statistical detection algorithm. Furthermore, by basing the detection criteria on the statistics of the nominal signal, the detection algorithm performs unsupervised, with no labelled anomalies necessary to distinguish process faults. Two statistical detection methods using a recursive autoregressive model (RAR) are presented in this work, the sequential probability ratio test (SPRT) and a RAR parameter-based statistic. These methods are verified on several naturally occurring and artificially induced test cases containing common process faults that appear in AM.