Abstract
V tem prispevku je predstavljena eksergijska analiza absorpcijskega hladilnega sistema, ki temelji na umetni inteligenci in deluje na principu hladilnega cikla litijev bromid–voda. Za absorpcijski hladilni sistem je značilno, da za svoje delovanje izkorišča srednjetlačno odjemno paro iz parne turbine. Eksergijska analiza umetne inteligence temelji na modelu strojnega učenja, ki napoveduje in optimizira delovanje absorpcijskega hladilnega sistema. Algoritem strojnega učenja je validiran z uporabo realnih procesnih podatkov. Rezultati kažejo, da absorpcijski hladilni sistem generira 126,71 kW hladu za daljinsko hlajenje in 279,57 kW toplote, ki se porabi za ogrevanje demineralizirane vode. V analiziranem obdobju je absorpcijski hladilni sistem v povprečju porabil 152,86 kW srednjetlačne pare in deloval s povprečnim eksergijskim izkoristkom 17,3 %. Študija nakazuje, da bi lahko eksergijski izkoristek hladilnega sistema izboljšali z uporabo manj kakovostne pogonske pare ali celo z uporabo vroče vode.
Keywords
absorption;analysis;artificial intelligence;cooling;efficiency;exergy;
Data
| Language: |
English |
| Year of publishing: |
2025 |
| Typology: |
1.01 - Original Scientific Article |
| Organization: |
UM FE - Faculty of Energetics |
| Publisher: |
Fakulteta za energetiko |
| UDC: |
620.9 |
| COBISS: |
239049731
|
| ISSN: |
1855-5748 |
| Views: |
0 |
| Downloads: |
5 |
| Average score: |
0 (0 votes) |
| Metadata: |
|
Other data
| Secondary language: |
Slovenian |
| Secondary title: |
Eksergijska analiza absorpcijskega hladilnega sistema z umetno inteligenco |
| Secondary abstract: |
An artificial intelligence (AI)-based exergy analysis of an absorption cooling system (ACS), utilizing a lithium bromide–water refrigeration cycle, is presented in this paper. The ACS is characterised by the utilisation of the intermediate-pressure (IP) extraction steam from the steam turbine for its operation. The exergy analysis of the ACS is detailed, based on AI modelling through a machine learning algorithm, which predicts and optimises the ACS performance. The machine learning algorithm is validated using real process data obtained through ACS measurements via the supervisory control and data acquisition (SCADA) system. The AI results show that the ACS generates 126.71 kW of cooling for district cooling and 279.57 kW of heat, which is used for heating demineralised water. During the analysis period, the ACS consumed an average of 152.86 kW of IP steam, and operated with an average exergy efficiency of 17.3%. The study suggests that the average exergy efficiency of the ACS could be improved by using lower-quality steam, or even hot water, for operation. |
| Secondary keywords: |
absorpcija;analiza;umetna inteligenca;hlajenje;učinkovitost;eksergija; |
| Type (COBISS): |
Scientific work |
| Pages: |
str. 21-34 |
| Volume: |
ǂVol. ǂ18 |
| Issue: |
ǂno. ǂ1 |
| Chronology: |
may 2025 |
| DOI: |
10.18690/jet.18.1.21-34.2025 |
| ID: |
26601065 |