Development of a prediction model for the electrode coating process in battery cell production

undergraduate thesis

• Title (croatian): Izrada prediktivnog modela procesa presvlačenja elektroda u proizvodnji baterijskih ćelija
• Author: Tin Samardžić
• Mentor: Marko Švaco (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2025-02-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

In this master's thesis, an in-depth overview of battery manufacturing, as well as the theory behind basic machine learning algorithms, was provided, with specific focus on the anode coating process. The data pipeline of BMW's Battery Cell Competence Center (BCCC) was also described in detail, particularly emphasizing the coating process. The primary goal of this work was to develop a machine learning model to predict pump speed, thereby optimizing the production of battery prototypes at the BCCC. To this end, data preparation was conducted to create an interpretable and usable dataset. The prediction model was divided into two parts: the Wet Loading Model and the Pump Speed Model. Each part was separately assessed, trained, and tested. Finally, the thesis introduced a data quality check script and the concept of DMC tracking.

Abstract (croatian)

U ovom diplomskom radu pružen je detaljan pregled proizvodnje baterija, kao i teorijska osnova osnovnih algoritama strojnog učenja, s posebnim naglaskom na proces nanošenja premaza na anodu. Također je detaljno opisan podatkovni tijek unutar BMW-ovog Battery Cell Competence Center (BCCC), s posebnim fokusom na proces premaza. Primarni cilj ovog rada bio je razviti model strojnog učenja za predviđanje brzine pumpe, čime bi se optimizirala proizvodnja prototipa baterija u BCCC-u. U tu svrhu provedena je priprema podataka kako bi se stvorio interpretabilan i upotrebljiv skup podataka. Prediktivni model podijeljen je u dva dijela: model mokrog opterećenja i model brzine pumpe. Svaki dio je zasebno procijenjen, treniran i testiran. Na kraju, rad uvodi skriptu za provjeru kvalitete podataka i koncept DMC praćenja.

• Keywords: BMW; battery production; coating process; machine learning; data analysis; data engineering; data quality check; DMC tracking
• Keywords (croatian): BMW; proizvodnja baterija; proces presvlačenja; strojno učenje; analiza podataka; inženjerstvo podataka; provjera kvalitete podataka; DMC praćenje
• Language: english
• URN:NBN: urn:nbn:hr:235:241749
• Study programme: Title: Mechanical Engineering; specializations in: Design, Process and Energy Engineering, Production Engineering, Engineering Modelling and Computer Simulation, Marine Engineering, Industrial Engineering and Management, Materials Engineering; Course: Engineering Modelling and Computer Simulation; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Embargoed access; Embargo expiration date: 2030-02-10
• Created on: 2025-02-07 19:29:56