CONVOLUTIONAL NEURAL NETWORK EVALUATION OF ADDITIVE MANUFACTURING IMAGES, AND ADDITIVE MANUFACTURING SYSTEM BASED THEREON

• Main Authors: GUERRIER, Paul, BAGGS, George
• Format: Patent
• Language: English
• Published: 25.03.2021
• Subjects: ADDITIVE MANUFACTURING TECHNOLOGY; ADDITIVE MANUFACTURING, i.e. MANUFACTURING OFTHREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVEAGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING,STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PERFORMING OPERATIONS; PHYSICS; REGULATING; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDEDFOR; SHAPING OR JOINING OF PLASTICS; TRANSPORTING; WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE, IN GENERAL

An additive manufacturing system uses a trained artificial intelligence module as part of a closed-loop control structure for adjusting the initial set of build parameters in-process to improve part quality. The closed-loop control structure includes a slow control loop taking into account in-process build layer images, and may include fast control loop taking into account melt pool monitoring data. The artificial intelligence module is trained using outputs from a plurality of convolutional neural networks (CNNs) tasked with evaluating build layer images captured in-process and images of finished parts captured post-process. The post process images may include two-dimensional images of sectioned finished parts and three-dimensional CAT scan images of finished parts.