diplomsko delo
Kristjan Cuznar (Author), Vito Logar (Mentor), Miha Glavan (Co-mentor)

Abstract

Diplomsko delo se osredotoča na problem izrabe zgodovinskih proizvodnih podatkov za namene izboljševanja kakovosti produktov. Obravnavan je proces hladnega valjanja, kjer je po obsežni digitalizaciji procesa na voljo podroben vpogled v procesne razmere. Hladno valjanje predstavlja enega pomembnejših postopkov pri izdelavi pločevine in je namenjeno zmanjšanju debeline, izenačitvi debeline ter zagotovitvi ustreznih mehanskih lastnosti obdelovanca. Da se zagotovi ustrezna kakovost izdelka, je izjemnega pomena ustrezna nastavitev valjavskega ogrodja, ki se običajno izvaja po receptih ter z ročnimi posegi operaterja pred in/ali med valjanjem. Pravila za korekcijo osnovnih receptov so običajno izkustvena, kar predstavlja znaten vpliv operaterja na končno kakovost izdelka. Za doseganje višje kakovosti izdelkov, večje konsistence pri obdelavi in zmanjšanje vpliva operaterjev, v diplomskem delu predlagamo podporno orodje, ki temelji na zgodovinskih podatkih o delovanju sistema in realno-časovnih meritvah procesnih veličin ter vsebuje ustrezne procesne modele, simulacijsko okolje in pravila, ki predlagajo ustreznejšo korekcijo parametrov recepta.

Keywords

optimizacija procesov;množični podatki;podatkovno rudarjenje;strojno učenje;modeliranje;identifikacija;odkrivanje znanja;visokošolski strokovni študij;Aplikativna elektrotehnika;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FE - Faculty of Electrical Engineering
Publisher: [K. Cuznar]
UDC: 004.8:681.5:621.77(043.2)
COBISS: 69050883 Link will open in a new window
Views: 633
Downloads: 196
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: Optimization of a cold-rolling process quality based on historical data
Secondary abstract: This thesis focuses on the problem of using historical production data for the purpose of product quality improvement. The subject of the discussion is cold rolling process where a detailed insight into the process conditions is available after an extensive digitalization of the process. Cold rolling is one of the most important processes in sheet metal production and is used for reducing the thickness, making thickness uniform, and ensuring the appropriate mechanical properties of the workpiece. In order to ensure the appropriate quality of the product, it is extremely important to adjust the rolling mill properly, which is set according to the recipes and with manual interventions of the operator before and/or during rolling. The rules for a base recipe correction are usually experiential, which shows a significant impact of the operator on the final product quality. For the purpose of achieving a higher product quality, a greater processing consistency, and a smaller influence of operators, the thesis proposes a support tool which is based on historical production data and real-time measurements of process variables. It also includes the appropriate process models, a simulation environment and the rules that suggest a more suitable base recipes correction.
Secondary keywords: process optimization;big data;data mining;machine learning;modelling;identification;knowledge extraction;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000315
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za elektrotehniko
Pages: XXIV, 110 str.
ID: 13089022