Additive manufacturing has the potential to revolutionize the production of complex and adapted parts, but is prone to manufacturing errors, ranging from minor inaccuracies and mechanical failures to complete failures associated with the production of rejects. It is therefore necessary to detect these deficiencies in a timely manner, to analyses and apply appropriate settings to minimize the possible occurrence of manufacturing errors. This article describes the proposed deep-learning monitoring system, which allows for a system of monitoring the quality of the printout during the process of additive manufacturing. The system identifies whether an error occurs during the printing process and may notify the operator if something went wrong. The combination of additive manufacturing, artificial intelligence, Raspberry Pi and online controls may create a comprehensive system for monitoring, managing and predicting process errors.
Science Journal 