CNCmachine tool digital twin applications with advancements of edge computing

Digital Twin application development for CNC machining processes is challenging since collecting real-time high-frequency process data from data sources such as dynamometer sensors and processing collected data into data infrastructures to develop analytical models without latency and time-synchronization is required. Dynamometers are used to measure cutting torque and cutting forces occur between the workpiece and cutting tool during machining process. Collecting dynamometer data during machining is expensive in terms of sensor mounting, data acquisition and processing. To overcome this challenge, a method for cutting torque estimation that utilizes edge computing application to collect internal machine data is proposed in this thesis, to establish preliminaries for machining Digital Twin applications with edge computing. The cutting torque is estimated during drilling experiments and prediction errors are calculated for experiments with same workpiece material as 0.31% and 0.9%, and with different workpiece material as 1.02% and 1.25%. Edge device allows the collection of high-frequency machining process data coming from spindle and feed-drive motors by providing inbuilt network components that enables data acquisition to Cloud sources from machine. Edge device signals are also presented in servo motor current and position closed-loop control loop block-diagrams to demonstrate data flow stages from CNC machine tool's internal servo system to Edge device. The block-diagrams are validated with Edge signals collected during milling and drilling experiments.