Details

Nastavitev PID z algoritmom diferencialne evolucije v PLK na primeru regulacije pretoka zraka
ID Bratina, Božidar (Author), ID Muškinja, Nenad (Author), ID Rotovnik, Milan (Author), ID Golob, Marjan (Author)

.pdfPDF - Presentation file, Download (637,17 KB)
MD5: E16249A96E12876C6CE98F1B65178076
URLURL - Source URL, Visit https://revija-ventil.si/wp-content/uploads/ventil-31-2025-2.pdf This link opens in a new window

Abstract
Prispevek predstavlja primer izvedbe naprednih algoritmov v PLK višjenivojskem programskem jeziku SCL za potrebe nastavitev parametrov PID na primeru procesne avtomatizacije. Z razvojem strojne opreme je možno napredne algoritme izvajati tudi direktno v PLK, še posebno zanimive pa so vgradnje algoritmov umetne inteligence in strojnega učenja. V okviru zaključnih del smo s študenti izvedli več prenosov naprednih algoritmov v PLK in jih primerjali s teoretičnimi metodami in praktičnimi orodji v industriji. Na področju uporabe optimizacijskih algoritmov v procesni industriji smo za namen demonstracije nastavitev parametrov PID izvedli z uporabo algoritma diferencialne evolucije na PLK na učnem laboratorijskem modelu regulacije pretoka zraka. Zaradi specifičnosti in omejenosti programskih jezikov so bile rešitve namenoma deloma izvedene tudi s pomočjo orodij generativne umetne inteligence, da se lahko ovrednoti kvaliteta podanih rešitev. Pojav generativnih modelov, kot so chatGPT, Copilot, Gemini, DeepSeek itd., nezadržno vpliva tudi na pedagoški proces in nove priložnosti podajanja učnih vsebin na vseh stopnjah izobraževanja.Izkušnje z različnimi projekti študentov avtomatike in robotike kažejo, da orodja umetne inteligence (UI) še ne zmorejo podati celovitih rešitev, lahko pa služijo kot dobra podpora.

Language:Slovenian
Keywords:procesna avtomatizacija, algoritmi vodenja, optimizacija parametrov PID, izobraževanje, orodja umetne inteligence
Work type:Article
Typology:1.01 - Original Scientific Article
Organization:FS - Faculty of Mechanical Engineering
Publication status:Published
Publication version:Version of Record
Year:2025
Number of pages:Str. 156-161
Numbering:Letn. 31, št. 3
PID:20.500.12556/RUL-178113 This link opens in a new window
UDC:681.5
ISSN on article:1318-7279
COBISS.SI-ID:242967043 This link opens in a new window
Publication date in RUL:19.01.2026
Views:282
Downloads:148
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Record is a part of a journal

Title:Ventil : revija za fluidno tehniko in avtomatizacijo
Shortened title:Ventil
Publisher:Univerza v Ljubljani, Fakulteta za strojništvo
ISSN:1318-7279
COBISS.SI-ID:54233856 This link opens in a new window

Licences

License:CC BY 4.0, Creative Commons Attribution 4.0 International
Link:http://creativecommons.org/licenses/by/4.0/
Description:This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.

Secondary language

Language:English
Title:PID tuning based on differential evolution algorithm in PLC for airflow control
Abstract:
The paper presents implementation of advanced algorithms in PLC using the higher-level programming language SCL for PID parameter tuning in process automation. Nowadays advanced algorithms can also be implemented directly in PLCs, whereas implementation of artificial intelligence and machine learning algorithms is particularly interesting. In scope of the master thesis and student projects we implemented several advanced algorithms in PLCs and compared them with theoretical methods and practical industry tools. In the study case we implemented differential evolution algorithm optimization algorithm in the process industry control scheme for PID parameter tuning into PLC for task of airflow control. Due to the specificity and limitations of programming languages, the solutions were partially developed using tools of generative artificial intelligence, to evaluate the quality of obtained solutions. Generative models such as chatGPT, Copilot, Gemini, DeepSeek, etc., also have an impact on the teaching part, bringing new opportunities for teaching content at all levels of education. Experience through various student projects by students of automation and robotics shows that AI tools are not yet able to provide comprehensive solutions, but they can serve as good support.

Keywords:process automation, control algorithms, PID parameter optimization, education, artificial intelligence tools

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Collection

This document is a part of these collections:
  1. Ventil

Back