Zaznavanje odstopanj v delovanju grelnikov jedrske elektrarne na podlagi regresijskih modelov : magistrsko delo

Obreza, Manja

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Zaznavanje odstopanj v delovanju grelnikov jedrske elektrarne na podlagi regresijskih modelov : magistrsko delo
Authors:	ID Obreza, Manja (Author) ID Igrec, Dalibor (Mentor) More about this mentor... ID Kelavić, Robert (Comentor)
Files:	MAG_Obreza_Manja_2026.pdf (2,45 MB) MD5: 6E3055557BFA4DE5FC539D15C29EC4ED
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FE - Faculty of Energy Technology
Abstract:	Magistrsko delo obravnava zaznavanje zgodnjih odstopanj v delovanju regenerativnih grelnikov napajalne vode v sekundarnem krogu Nuklearne elektrarne Krško. Na podlagi razpoložljivih procesnih meritev je razvit empirični regresijski model, ki opisuje pričakovano obnašanje grelnika med normalnim obratovanjem. Odstopanja se identificirajo z rezidualno analizo, s katero se namesto statičnih alarmnih pragov spremlja neskladje med napovedjo modela in meritvijo v danih obratovalnih pogojih. Poseben poudarek je na predobdelavi podatkov, izboru vhodnih spremenljivk in validaciji modela s statističnimi kazalniki. Prag detekcije je določen objektivno z ROC-analizo in Youdenovim kriterijem. Učinkovitost metode je prikazana na dveh scenarijih s simulirano napako, dvigu temperature drenaže in dvigu pretoka drenaže grelnika 6A.
Keywords:	zaznavanje odstopanj, grelnik napajalne vode, linearni regresijski model, rezidualna analiza, Nuklearna elektrarna Krško
Place of publishing:	Krško
Place of performance:	Krško
Publisher:	[M. Obreza]
Year of publishing:	2026
Number of pages:	XVII, 74 str.
PID:	20.500.12556/DKUM-97666
UDC:	621.311.25:621.1.016.4(043.2)
COBISS.SI-ID:	280676611
Publication date in DKUM:	05.06.2026
Views:	200
Downloads:	39
Metadata:
Categories:	FE
:	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:	31.03.2026

Secondary language

Language:	English
Title:	Detection of deviations in the operation of nuclear power plant heaters based on regression models
Abstract:	This master’s thesis addresses early detection of abnormal behaviour in regenerative feedwater heaters in the secondary cycle of the Krško Nuclear Power Plant. Using available process measurements, a data-driven regression model is developed to describe the heater’s expected behaviour under normal operating conditions. Abnormal conditions are identified through residual analysis, where the difference between the model prediction and the actual measurement is tracked under the same operating conditions. The work focuses on data preprocessing, selecting key input variables, and model validation using standard statistical indicators. Instead of fixed alarm limits, the detection threshold is set objectively using ROC analysis and the Youden criterion. The approach is demonstrated on two synthetic fault scenarios for Heater 6A: an increase in drain temperature and an increase in drain flow.
Keywords:	anomaly detection, feedwater heater, linear regression model, residual analysis, Krško Nuclear Power Plant

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back