Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1267936

Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit


Ujević Andrijić, Željka; Rimac, Nikola
Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit // Book of Abstracts / Rogošić, Marko (ur.).
Zagreb: Hrvatsko društvo kemijskih inženjera i tehnologa (HDKI), 2023. str. 176-176 (poster, međunarodna recenzija, sažetak, znanstveni)


CROSBI ID: 1267936 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit

Autori
Ujević Andrijić, Željka ; Rimac, Nikola

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Book of Abstracts / Rogošić, Marko - Zagreb : Hrvatsko društvo kemijskih inženjera i tehnologa (HDKI), 2023, 176-176

Skup
28th Croatian Meeting of Chemists and Chemical Engineers

Mjesto i datum
Rovinj, Hrvatska, 28.03.2023. - 31.03.2023

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
NTU ; Neural Networks ; Crude Distillation Unit

Sažetak
Monitoring of fouling deposition inside crude distillation unit preheat train is a serious challenge due to the lack of analyzed physicochemical variables of crude oils, unknown mechanisms of deposit formation, and large deviations from designed process conditions. In addition, real plant processes operate under dynamic conditions with frequent changes in the process regime. Fouling, or the formation of deposits in the preheat train of CDU, is an operational problem with serious consequences for oil refineries, with half of the refinery's energy costs attributable to the crude oil preheater. [1] These unfavorable conditions complicate the development of first-principle models. It has become a common practice in recent years to use data-driven models like neural networks (NN) instead of empirical or fundamental models for fouling detection. [2] However, simpler monitoring should be possible by using design equations, and by following heat exchangers effectiveness and NTU correlations. The NTU parameter summarizes geometric (via the transfer area) and hydrodynamic properties (via the overall heat transfer coefficient and the sum of fluid flow rate and heat capacity). [3] This property provides a useful basis for the heat exchanger performance model. In this study, a comparison is made between the NTU and NN models for heat exchangers in the preheating train of a crude oil distillation unit. The results show that although the NTU model is less accurate, it indicates a gradual increase in fouling factor during the time from overhaul to process stop, which provides useful information to process engineers about the fouling dynamics and effectiveness of the heat exchanger.

Izvorni jezik
Engleski

Znanstvena područja
Kemijsko inženjerstvo



POVEZANOST RADA


Ustanove:
Fakultet kemijskog inženjerstva i tehnologije, Zagreb

Profili:

Avatar Url Željka Ujević Andrijić (autor)

Avatar Url Nikola Rimac (autor)


Citiraj ovu publikaciju:

Ujević Andrijić, Željka; Rimac, Nikola
Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit // Book of Abstracts / Rogošić, Marko (ur.).
Zagreb: Hrvatsko društvo kemijskih inženjera i tehnologa (HDKI), 2023. str. 176-176 (poster, međunarodna recenzija, sažetak, znanstveni)
Ujević Andrijić, Ž. & Rimac, N. (2023) Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit. U: Rogošić, M. (ur.)Book of Abstracts.
@article{article, author = {Ujevi\'{c} Andriji\'{c}, \v{Z}eljka and Rimac, Nikola}, editor = {Rogo\v{s}i\'{c}, M.}, year = {2023}, pages = {176-176}, keywords = {NTU, Neural Networks, Crude Distillation Unit}, title = {Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit}, keyword = {NTU, Neural Networks, Crude Distillation Unit}, publisher = {Hrvatsko dru\v{s}tvo kemijskih in\v{z}enjera i tehnologa (HDKI)}, publisherplace = {Rovinj, Hrvatska} }
@article{article, author = {Ujevi\'{c} Andriji\'{c}, \v{Z}eljka and Rimac, Nikola}, editor = {Rogo\v{s}i\'{c}, M.}, year = {2023}, pages = {176-176}, keywords = {NTU, Neural Networks, Crude Distillation Unit}, title = {Comparison of NTU and neural network models for fouling monitoring of heat exchangers in the preheat train section of a Crude Distillation Unit}, keyword = {NTU, Neural Networks, Crude Distillation Unit}, publisher = {Hrvatsko dru\v{s}tvo kemijskih in\v{z}enjera i tehnologa (HDKI)}, publisherplace = {Rovinj, Hrvatska} }




Contrast
Increase Font
Decrease Font
Dyslexic Font