Cold extrusion of components (also called cold forging and cold impact extrusion) requires a sequence of operations to transform an initial simply shaped billet into a desired complex part. To determine the most feasible extrusion sequence is an intensive knowledge-based process planning task currently done by experienced process designers. In this study, a prototype of a knowledge-based system is developed which can become a very useful aide to human process planners. In the system, parts are classified into groups and a features representation is used. A new classification of extrusion operations is also introduced. To generate process plans, the system selects operators corresponding to the reverse of actual extrusion operations. The feasibility of each selected operator is checked with empirical rules. The strategy of backward exhaustive search for operators is employed. The current system requires input of the geometry of a part to be produced along with certain material and process data. Then the formable part geometry and process plans are generated automatically. The system also rates the process plans from the perspective of achieving economical production. A cost estimating model is proposed so that users can obtain production cost for a selected process plan based on the data they provide.
| Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:dissertations-8249 |
| Date | 01 January 1992 |
| Creators | Sun, Qiang |
| Publisher | ScholarWorks@UMass Amherst |
| Source Sets | University of Massachusetts, Amherst |
| Language | English |
| Detected Language | English |
| Type | text |
| Source | Doctoral Dissertations Available from Proquest |
Page generated in 0.0018 seconds