Monitoring temperature in additive manufacturing with physics-based compressive sensing

Sensing is one of the most important components in manufacturing systems to ensure the high quality of products. However, the deployment of a large number of sensors increases the costs of manufacturing systems for both operation and maintenance. Processing the large amount of sensor data for real-time process monitoring is also challenging. Recently compressive sampling or compressed sensing (CS) approaches have been developed to reduce the amount of data collection. However, the reduction is limited to individual sensor types and compression ratio is not high. In this paper, a physics-based compressive sensing (PBCS) approach is proposed to improve the traditional CS approach based on the physical knowledge of phenomenon in applications. The volume of data and the number of sensors needed for process monitoring are significantly reduced. This approach is applied to monitor the temperature field of additive manufacturing processes. In the experimental study, only a few number of thermal readings are needed to reconstruct the complete three-dimensional temperature field using the PBCS approach.