Design space exploration in engineering automation
| dc.contributor.advisor | Rehof, Jakob | |
| dc.contributor.author | Chaumet, Constantin | |
| dc.contributor.referee | Meyer, Anne | |
| dc.date.accepted | 2025-09-22 | |
| dc.date.accessioned | 2025-10-20T08:39:59Z | |
| dc.date.available | 2025-10-20T08:39:59Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | Two approaches that enable design space exploration and improve the automation of engineering tasks are developed. The first approach leverages combinatory logic synthesis with intersection types and predicates as an instrument to persist domain-specific knowledge held by CAD designers, i.e., how individual parts connect and can be composed, to automatically generate and assemble designs in CAD software. This enables an iterative exploration and reduction of the design space to relevant results, based on a set of performance metrics and structural constraints. The approach speeds up the creation of CAD assemblies while simultaneously improving quality by ensuring that assembly trees are uniformly organized and no joints are omitted. A set of parts suitable for synthesizing robotic arms is presented and used to perform a case study evaluating the practical applicability of the design space exploration and the overall performance compared to a human user. The second approach enables the automatic generation of form-fitting mold jaws for parts with complex geometry. This approach allows these parts, intended to be additively manufactured, to be gripped by a vise for post-processing. This provides a rigid clamping solution that is easy to operate and reduces the necessary clamping force. The generated mold jaws may accommodate multiple different parts and/or the same part in different alignments. The approach can also ensure that geometry remains accessible to the machining tool and can account for manufacturing tolerances. The automatic generation of mold jaws is utilized to implement an optimization loop that determines Pareto-optimal rotations of each part or alignment around the cutting tool's axis with respect to clamping-induced deformation. Experimental validation is performed. The software aspects of the approach are validated by utilizing a topology-optimized part, whose optimal rotations are known a priori. The technological aspects are validated by manufacturing the part through selective laser sintering and inspecting the surface-to-surface contact area during clamping with a pressure-sensitive film. | en |
| dc.identifier.uri | http://hdl.handle.net/2003/44037 | |
| dc.identifier.uri | http://dx.doi.org/10.17877/DE290R-25805 | |
| dc.language.iso | en | |
| dc.subject | Design space exploration | en |
| dc.subject | Combinatory logic synthesis | en |
| dc.subject | Additive manufacturing | en |
| dc.subject | Post-proccessing | en |
| dc.subject | Machining | en |
| dc.subject | Clamping | en |
| dc.subject | Optimization | en |
| dc.subject.ddc | 004 | |
| dc.subject.rswk | Logiksynthese | de |
| dc.subject.rswk | Kombinatorische Logik | de |
| dc.subject.rswk | Nachbearbeitung | de |
| dc.subject.rswk | Optimierung | de |
| dc.title | Design space exploration in engineering automation | en |
| dc.type | Text | |
| dc.type.publicationtype | PhDThesis | |
| dcterms.accessRights | open access | |
| eldorado.dnb.deposit | true | |
| eldorado.secondarypublication | false |