Hibridno modeliranje kontinuirnega mešalnega reaktorja z nevronskimi mrežami

Jurkovnik, Jonatan

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Hibridno modeliranje kontinuirnega mešalnega reaktorja z nevronskimi mrežami
Authors:	ID Jurkovnik, Jonatan (Author) ID Novak Pintarič, Zorka (Mentor) More about this mentor... ID Pečar, Darja (Comentor)
Files:	UN_Jurkovnik_Jonatan_2026.pdf (2,90 MB) MD5: 33E8D4F077F6443D26441A2988072940
Language:	Slovenian
Work type:	Bachelor thesis/paper
Typology:	2.11 - Undergraduate Thesis
Organization:	FKKT - Faculty of Chemistry and Chemical Engineering
Abstract:	V diplomski nalogi smo v programskem jeziku Python modelirali kontinuirni mešalni reaktor (CSTR) z uporabo mehanističnega in hibridnega modela. Mehanistični model ocenjuje koncentracije z uporabo eksaktnih zakonov kemije in fizike. Hibridni model združuje mehanistično modeliranje s podatkovno vodenim, ki koncentracije računa z uporabo nevronskih mrež. V našem primeru je mehanistični del hibridnega modela izračunaval spremembe koncentracij komponente zaradi vtokov in iztokov komponent, nevronska mreža pa je ocenjevala hitrost nastanka produkta in porabe reaktantov zaradi kemijske reakcije. Pri mehanističnem modelu smo hitrost zaradi kemijske reakcije ocenili z uporabo kinetične enačbe. Hibridni model se je izkazal kot uporaben, saj se njegovi rezultati dobro ujemajo z rezultati mehanističnega modela. Z nastavitvijo parametrov nevronske mreže lahko vplivamo na kompromis med natančnostjo modela in računskim časom.
Keywords:	nevronske mreže, kontinuirni mešalni reaktor, modeliranje, kemijska reakcija, strojno učenje, kemijsko inženirstvo
Place of publishing:	Maribor
Year of publishing:	2026
PID:	20.500.12556/DKUM-99326
Publication date in DKUM:	07.09.2026
Views:	219
Downloads:	9
Metadata:
Categories:	KTFMB - FKKT
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	12.08.2026

Secondary language

Language:	English
Title:	Hybrid modeling of a continuous stirred tank reactor using neural networks
Abstract:	In our thesis, we modelled a continuous stirred tank reactor (CSTR) in Python using both mechanistic and hybrid models. The mechanistic model estimates concentrations using the exact laws of chemistry and physics. The hybrid model combines mechanistic and data-driven modelling, calculating concentrations using neural networks. In our case, the mechanistic part of the hybrid model calculated changes in component concentrations due to inflows and outflows, while the neural network estimated the rates of product formation and reactant consumption due to a chemical reaction. In the purely mechanistic model, we estimated the reaction rate using a kinetic equation. The hybrid model proved useful, as its results agreed well with those of the mechanistic model. By adjusting the parameters of the neural network, we can control the trade-off between model accuracy and computational time.
Keywords:	neural networks, continuous stirred tank reactor, modeling, chemical reaction, machine learning, chemical engineering

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back