Neural Network Based Quality Increase Of Surface Roughness Results In Free Form Machining

Abstract

This paper concerns with free form surface reorganization and assessment of free form model complexity, grouping particular surface geometrical properties within patch boundaries, using self organized Kohonen neural network (SOKN). Neural network proved itself as an adequate tool for considering all topological non-linearities appearing in free form surfaces. Coordinate values of point cloud distributed at a particular surface were used as a surface property's descriptor, which was led into SOKN where representative neurons for curvature, slope and spatial surface properties were established. On a basis of this approach, surface patch boundaries were reorganized in such a manner that finish machining strategies gave best possible surface roughness results. The patch boundaries were constructed regarding to the Gaussian and mean curvature, in order to achieve smooth transition between patches, and in this way preserve or even improve desired curve and surface continuities, (C2 and G2). It is shown that by reorganization of boundaries considering curvature, slope and spatial point distribution, the surface quality of machined free form surface is improved. Approach was experimentally verified on 22 free form surface models which were reorganized by SOKN and machined with finish milling tool-path strategies. Results showed rather good improvement of mean surface roughness profile Ra for reorganized surfaces, when comparing to unorganized free form surfaces.

Ovaj se članak bavi reorganizacijom slobodnih površina i ocjenom kompleksnosti modela slobodnih površina, grupirajući određena površinska geometrijska svojstva unutar zatvorenih površina, koristeći samoorga-nizirajuću Kohonenovu neuronsku mrežu (SOKN). Neuronske mreže pokazale su se kao prikladan alat za razmatranje svih topoloških nelinearnosti koje se pojavljuju kod slobodnih površina. Vrijednosti koordinata oblaka točaka raspodijeljenih nad određenom površinom korišteni su kao svojstveni opis, što je nadalje vodilo prema SOKN-u, gdje su ustanovljeni reprezentativni neuroni za zakrivljenost, nagib i prostorno-površinska svojstva. Na temelju ovoga pristupa reorganizirane su granice zatvorenih površina na takav način da metode površinske obrade daju najbolje moguće rezultate spram površinske hrapavosti. Granice tih površina određene su prema Gaussovoj i prosječnoj zakrivljenosti kako bi se postigao glatki prijelaz između zatvorenih površina te kako bi se na taj način očuvala ili čak unaprijedila željena zakrivljenost i glatkoća površine, (C2 i G2). Pokazano je da se reorganizacijom granica s obzirom na zakrivljenost, nagib i prostornu raspodjelu točaka, poboljšava kvaliteta obrađene slobodne površine. Pristup je eksperimentalno potvrđen na 22 modela slobodne površine koji su reorganizirani SOKN-om i površinski obrađeni određenim metodama. Rezultati pokazuju poprilično dobro poboljšanje prosječne vrijednosti površinske hrapavosti Ra za reorganizirane strukture u usporedbi sa neorganiziranim slobodnim površinama.

Key words:

• Neural network (NN)
• Self organized Kohonen neural network (SOKN)
• Free form surface
• place-CAM
• Index of surface complexity (ISC)

Ključne riječi:

• neuronske mreže
• samoorganizirajuća Kohonen neuronska mreža (SOKN)
• slobodna površina
• placeCAM
• indeks površinske kompleksnosti

Additional information

Notes on contributors

Marjan Korošec

Marjan Korošec received his Master degree in 1997 in Faculty of Mechanical engineering—Ljubljana in an area of of Automation of production systems. From 1998 to 1999 he was employed as an assistant in metal cutting Laboratory in Faculty of Mechanical Engineering in Ljubljana. From 1999 to 2001 he was a leader of research and development group in tool—shop company Saturnus—Ljubljana. From 2001 to 2003 he was an independent entrepreneur in an area of representation of CAD/CAM systems and solutions of environmental problematics. He received his Ph.D. degree in 2003 in Faculty of Mechanical Engineering Maribor, in an area of intelligent machining He works as an assistant in Laboratory for Computer added design—Lecad in Faculty of Mechanical engineering Ljubljana.