magistrsko delo
Beno Jurjovec (Author), Jože Balič (Mentor), Miha Kovačič (Co-mentor)

Abstract

V magistrskem delu je predstavljeno napovedovanje izmeta jeklenih valjancev po pregledu na kontrolni liniji. Osredotočili smo se na izmet zaradi površinskih napak, na luščenih okroglih valjancih, pri kvaliteti 30MnVS6. Beležili smo kemično sestavo taline, toplotni tok, hitrost litja med odlivanjem jekla na trožilni napravi za kontinuirano odlivanje jekla in procent izmeta zaradi površinskih napak, v obdobju od septembra 2014 do maja 2015. Na podlagi zbranih podatkov sta bila izdelana modela s pomočjo linearne regresije in genetskega programiranja. Model za napovedovanje izmeta s pomočjo sistema za genetsko programiranje je 1,57-krat boljši od modela dobljenega s pomočjo linearne regresije. Izsledki raziskave so v praksi uporabljeni od sredine leta 2015. Izmet je pri kvaliteti 30MnVS6 za 3,09-krat manjši. Tako znaša letni prihranek, pri količini 12.000 t jeklenih valjancev iz 30MnVS6, 460.000 €.

Keywords

površinske napake na valjancih;modeliranje;linearna regresija;genetsko programiranje;napoved izmeta;valjano jeklo;magistrske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UM FS - Faculty of Mechanical Engineering
Publisher: [B. Jurojovec]
UDC: 004.89:[620.191:669.14-122](043.2)
COBISS: 20177430 Link will open in a new window
Views: 1072
Downloads: 107
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: STEELMAKING PROCESSES IMPACT ON 30MNVS6 AND OCCURRENCE OF SURFACE DEFECTS WITH THE USE OF GENETIC PROGRAMMING
Secondary abstract: In this master thesis predicting of rolled steel scrap using control line results is presented. The thesis focuses on the analysis of scrap due to surface defects on peeled round bars made from 30MnVS6 steel grade. Chemical composition of the melt, heat flux, speed casting during casting, using three strands continuous caster, and scrap percentage due to surface defects were monitored from September 2014 till May 2015. According to collected data the models using linear regression and genetic programming were obtained. Genetic programming model for predicting scrap percentage is 1.57-times better than the model obtained using linear regression. The results of the research are used in practice from middle of 2015. The 30MnVS6 scrap percentage is 3.09-times lower. Accordingly, the annual savings, at 12,000 t of 30MnVS6 production, amount 460,000 EUR.
Secondary keywords: surface defects in rolling;molding;linear regression;genetic programming;forecast ejection;rolled steel;
URN: URN:SI:UM:
Type (COBISS): Master's thesis
Thesis comment: Univ. v Mariboru, Fak. za strojništvo
Pages: VI, 50 str.
ID: 9156570