Uporaba strojnega vida za zagotavljanje kakovosti proizvodnje zaganjalnikov

magistrsko delo magistrskega študijskega programa II. stopnje Strojništvo

Gregor Gergolet (Author), Davorin Kramar (Mentor), Andrej Lebar (Co-mentor)

Abstract

Namen magistrske naloge je kontrola premera obročev in višine vilic na paletah z uporabo strojnega vida. Z metodo Poka Yoke odpravimo napake, ki jih povzroča človeški faktor in preprečimo njihov negativni vpliv na končni proizvod. Pri nadzoru procesa s kamerami imamo težave pri detekciji premera obročev in višine vilic. Obrabljeno orodje in odbojni paneli vplivajo na slabo detekcijo geometrijskih likov s kamero zaradi različnih odsevov. Z uporabo metod in orodij kakovosti smo analizirali trenutno stanje in ovrednotili najbolj kritične napake. Izvedli smo načrt za izboljšanje stabilnosti sistema. Z uporabo drugih odbojnih panelov, vgradnje polarizacijskega filtra in pravilne postavitve kamer, smo dosegli bolj natančne rezultate pri detekciji pravilno izbranega orodja. Primerjali smo prvotno in končno stanje. Ugotovili smo, da z nadgradnjo sistema pri merjenju višine vilic in premera obročev dosegli bistveno boljše rezultate. Izdelali smo FMEA analizo prvotnega in končnega stanja. Iz analize FMEA smo ugotovili, da se je faktor prednostnega števila tveganja v primerjavi s prvotnim stanjem bistveno znižal.

Keywords

magistrske naloge;strojni vid;detekcija robov s kamerami;metode in orodja zagotavljanja kakovosti;Poka Yoke;FMEA;

Data

Language:	Slovenian
Year of publishing:	2018
Typology:	2.09 - Master's Thesis
Organization:	UL FS - Faculty of Mechanical Engineering
Publisher:	[G. Gergolet]
UDC:	004.946:681.518.52(043.2)
COBISS:	16396059
Views:	1534
Downloads:	872
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Using vison sensors to ensure quality on starter motors' assembly line
Secondary abstract:	The objective of this master thesis was to ensure control of the ring diameter and forks height on pallets with the support of vision sensors. With the Poka Yoke method we were able to get rid of human error, which had an impact on the final product. The process monitoring process using the cameras revealed issues in measuring the diameter of the rings and the height of the forks. Firstly, by using methods and quality tools the current status in the pallet rearrangement working space was analysed. Furthermore, using reflective panels, a polarised filter and the appropriate cameras position, more accurate results were achieved. A final FMEA analysis was performed comparing the current and previous set up. This revealed a significant reduction of the RPN factor.
Secondary keywords:	vision camera;detection borders with cameras;methods and tools of quality check;Poka Yoke;FMEA;
Type (COBISS):	Master's thesis/paper
Study programme:	0
Embargo end date (OpenAIRE):	1970-01-01
Thesis comment:	Univ. Ljubljana, Fak. za strojništvo
Pages:	XXII, 75 str.
ID:	10957887