| Περίληψη: | The use of industrial robots for machining operations has attracted the attention of many industries lately, as they are able to significantly increase the production system flexibility and reduce the production cost, while present several advantages over conventional CNC machines. However, several challenges have slowed down the adoption of robots in machining tasks by industry. Industrial robots are constructed as serial, open kinematic chains, leading to insufficient structural stiffness, which in turn directly affects the machining process accuracy. In addition to this, due to the posture dependency of the robot dynamic behavior, time consuming and costly experiments are required to determine its dynamic behavior over its whole working space with traditional approaches. For this purpose, this work aims to develop a complete Digital-Model for robotic machining that can be used during the process planning stage, providing a tool for virtual commissioning of the process. For the robot arm modelling, the flexible links-flexible joints approach has been selected and the Multi-Body Simulation method combined with a Component Mode synthesis method have been adopted for the simulation of its dynamic behavior. The cutting tool deflections due to robot deformation during the process can be calculated with this model too. Additionally, motivated from the tremendous flexibility and versatility of robotic based machining systems two optimization algorithms have been developed, in an effort to increase process accuracy and efficiency. Specifically, an algorithm that calculates the workpiece optimal placement position with respect to robot stiffness has been developed, acquiring knowledge from the robot stiffness maps we generated using the above-mentioned model. Also, an algorithm calculating the optimal feed-rate of the robot arm during the machining process has been developed, attempting to constraint the contour error by regulating the generating cutting forces without affecting the productivity. Finally, the developed model and algorithms have been applied in two different case study parts to present their capabilities and functionality. Results showed that with the application of the two developed optimization algorithms the process accuracy can be improved up to 45%, compared to the unoptimized process setup.
|
|---|
