diplomsko delo
Abstract
S pomočjo algoritmov je mogoče na pretočnem reaktorju narediti platformo
za avtomatizirano optimizacijo obratovalnih pogojev kemijskih reakcij za doseg želenih
rezultatov, kot so dobitek, učinkovitost, cena produkta ipd. Optimizacija deluje na
osnovi povratne informacijske zanke med reakcijskim sistemom in algoritmom, pri
kateri algoritem spreminja obratovalne pogoje. Posledično je rezultat reakcije drugačen
in na podlagi tega, kako se le-ta spreminja, algoritem dobiva nove podatke, na podlagi
katerih določa nadaljnje korake pri iskanju optimalnih pogojev.
Keywords
pretočni sistemi;optimizacija;strojno učenje;algoritmi;diplomska dela;
Data
Language: |
Slovenian |
Year of publishing: |
2022 |
Typology: |
2.11 - Undergraduate Thesis |
Organization: |
UL FKKT - Faculty of Chemistry and Chemical Technology |
Publisher: |
[M. Levičar] |
UDC: |
66.023.2(043.2) |
COBISS: |
124471299
|
Views: |
27 |
Downloads: |
11 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
Self-optimisation of chemical reactions in flow systems |
Secondary abstract: |
With the help of algorithms, it is possible to turn a flow reactor into a
platform for optimizing reaction conditions of chemical reactions to achieve the desired
results, such as yield, efficiency, product cost, etc. Optimization is based on a feedback
loop between the reaction system and the algorithm where the algorithm changes
reaction conditions. Consequently, the result of the reaction is different. Based on how it
changes, the algorithm receives new data from which it determines further steps in
finding the optimal conditions. |
Secondary keywords: |
flow systems;reactions;machine learning;algorithms;Kemijske reakcije;Univerzitetna in visokošolska dela; |
Type (COBISS): |
Bachelor thesis/paper |
Study programme: |
1000372 |
Embargo end date (OpenAIRE): |
1970-01-01 |
Thesis comment: |
Univ. v Ljubljani, Fak. za kemijo in kemijsko tehnologijo, UNI Kemijsko inženirstvo |
Pages: |
25 str. |
ID: |
16336550 |