Automatic Process Planning For a Five-Axis Additive Hybrid Manufacturing System

Open Access
- Author:
- Xiao, Xinyi
- Graduate Program:
- Industrial Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- July 09, 2020
- Committee Members:
- Sanjay Joshi, Dissertation Advisor/Co-Advisor
Sanjay Joshi, Committee Chair/Co-Chair
Timothy W. Simpson, Committee Member
Edward William Reutzel, Outside Member
Vittaldas V Prabhu, Committee Member
Robert Carl Voigt, Program Head/Chair - Keywords:
- Hybrid Manufacturing
Process Planning
Automation
Additive Manufacturing - Abstract:
- A multi-axis additive manufacturing (AM) system allows for reorienting of the geometry during a build to gain greater building flexibility over that of traditional planar layer by layer additive manufacturing processes. A hybrid manufacturing (HM) system integrates computer numerical control (CNC) machining with multi-axis AM into one process that can switch between each of these two processes, reaping the benefits of both. Currently, these two systems require significant manual work to transform the CAD design into a manufactured part. With the lack of automated process planning algorithms to avoid the significant amount manual work necessary, adoption of HM technology has been slow. Critical components of process planning in multi-axis AM and HM include: 3D model decomposition, sequencing of production of the decomposed volumes, and toolpath generation. Three process planning approaches are presented in this dissertation which seek to reduce the manual work required by automating each of the critical components. The first two approaches rely on the concept of generating decomposed volumes that are self-supported, and sequencing these volumes in a manner that avoids collisions between the build and the AM or HM system, then mapping tool path strategies to each of these volumes. The first approach treats the five-axis machine as a 3+2 axis machine, where the rotational axes are only used for positioning and the decomposed volumes only accommodate planar tool paths. The 2nd approach uses the full five-axis capability and 3D tool paths are used to decompose the part into self supported volumes that can be built without additional support structures. The 3rd approach, referred to as direct five-axis slicing, eliminates the volume decomposition and can directly generate the 3D slices, associating each slice with a tool path. All the approaches are focused on eliminating support structures and avoiding local collision between the tool and the part. Algorithms for decomposition are developed based on the process of identifying concave edges in a part’s geometry and segmenting the part along these edges using the surfaces generated by the concave edges. For each decomposed volume, a build direction is identified along with the building sequence and toolpath strategy that can be used to generate the detailed toolpaths. Several case studies using the developed algorithms are presented, along with simulations and experimental results, to validate and showcase the capabilities of the three proposed process planning approaches. A comparison of the three approaches is also included to highlight the features and the limitations of each approach.