Details

Razvoj tehnologij inkrementalnega preoblikovanja na visoko-energijskih preoblikovalnih strojih
ID Sevšek, Luka (Author), ID Pepelnjak, Tomaž (Mentor) More about this mentor... This link opens in a new window, ID Šimic, Marko (Comentor)

.pdfPDF - Presentation file, Download (6,87 MB)
MD5: 437845210BE4E311B0C5376AD8DC4C9C

Abstract
Pretirano tanjšanje pločevine in neenakomerna porazdelitev debeline stene končnega izdelka je ena glavnih slabosti enotočkovnega inkrementalnega preoblikovanja SPIF. To značilnost postopka SPIF lahko izboljšamo z nadgradnjo v hibridno dvostopenjsko preoblikovanje, sestavljeno iz zaporedne izvedbe izbočevanja in SPIF. Poglobljena analiza vplivnih parametrov postopka SPIF in dodatnega izbočevanja z uporabo metode naključnih gozdov je izpostavila najbolj vplivne parametre. Dodatno je bilo moč z metodo umetne nevronske mreže in genetskega algoritma opredeliti procesno okno tehnoloških pogojev, geometrijskih veličin orodja in trajektorij gibanja orodja, pri katerih dobimo pri zaporedju izbočevanja ter postopka SPIF enakomernejšo porazdelitev debeline stene izdelka v primerjavi s postopkom SPIF. Z namenom izvajati SPIF na težje preoblikovalnih jeklih, vključno z dvofaznimi (DP) jekli, je bil eden glavnih ciljev doktorske naloge razvoj hibridnega stroja, ki temelji na hidravlični stiskalnici. Ker se pehalo hidravlične stiskalnice pomika samo v vertikalni smeri, sta bili dve dodatni osi gibanja, potrebni za izvedbo SPIF, doseženi z namestitvijo linearnih vodil in pogonov na mizo stiskalnice. Dodatno je bilo hibridno dvostopenjsko preoblikovanje izvedeno brez prepenjanja preoblikovanca med obema stopnjama preoblikovanja, kar privede do večje produktivnosti in večje natančnosti preoblikovanja z eliminacijo človeškega vpliva in vpliva prepenjanja.

Language:Slovenian
Keywords:enotočkovno inkrementalno preoblikovanje, hibridno dvostopenjsko preoblikovanje, umetna nevronska mreža, genetski algoritem, hidravlična stiskalnica
Work type:Doctoral dissertation
Typology:2.08 - Doctoral Dissertation
Organization:FS - Faculty of Mechanical Engineering
Year:2025
Number of pages:XXVI, 166 str.
PID:20.500.12556/RUL-174573 This link opens in a new window
UDC:621.9.044:621.226:004.032.26(043.3)
COBISS.SI-ID:251912963 This link opens in a new window
Publication date in RUL:04.10.2025
Views:506
Downloads:200
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Development of incremental forming technologies on high-energy forming machines
Abstract:
Excessive sheet thinning and uneven wall thickness distribution of the final product is one of the main disadvantages of single-point incremental forming SPIF. This characteristic of the SPIF process can be improved by extending the mentioned process to a hybrid two-step process consisting of a sequential implementation of bulging and SPIF. An in-depth influencial parameter analysis of SPIF process and the additional bulging using the random forest method resulted in highlighting most influential parameters. In addition, using the artificial neural network method and genetic algorithm, it was possible to define a process window of technological conditions, geometric sizes of tools and trajectories of tool movement, in which a more uniform distribution of the wall thickness of the product is achieved in the sequence of bulging and SPIF process compared to SPIF. To perform SPIF on steels that are more difficult to form, including dual-phase (DP) steels, one of the main objectives of the doctoral thesis was to develop a hybrid machine based on a hydraulic press. As the hydraulic press ram only moves in the vertical direction, the two additional axes of movement required to perform SPIF were achieved by installing linear guides and drives on the press table. In addition, the hybrid two-step forming was carried out without reclamping the workpiece between the two forming steps, which resulted in higher productivity and greater forming accuracy, as both the human influence and the influence of reclamping were eliminated.

Keywords:single-point incremental forming, hybrid two-step forming, artificial neural network, genetic algorithm, hydraulic press

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back