Detekcija anomalij v meritvah onesnaževal zraka

diplomsko delo

Luka Volk (Author), Petar Vračar (Mentor), Jana Faganeli Pucer (Co-mentor)

Abstract

Onesnaženost zraka predstavlja resen okoljski problem, ki lahko slabo vpliva na zdravje ljudi in kakovost okolja. V diplomski nalogi se osredotočamo na detekcijo anomalij v meritvah onesnaževal zraka: žveplovega dioksida (SO2), ozona (O3), dušikovega dioksida (NO2), dušikovega oksida (NO), ogljikovega monoksida (CO) in delcev PM10. Podatki so bili pridobljeni s strani Agencije Republike Slovenije za okolje (ARSO). Zaznavanje anomalij v meritvah je ključno za zagotavljanje zanesljivih podatkov, saj lahko anomalije v podatkih kažejo na tehnične napake senzorjev oziroma kakšen drug izreden dogodek. V nalogi smo implementirali in primerjali tri različne modele strojnega učenja: XGBoost, LSTM samokodirnik in matrični profil. Med analiziranimi metodami se je najbolje izkazal model XGBoost, saj je uspešno zaznal največje število anomalij ter dosegel najvišje vrednosti metrik za ocenjevanje uspešnosti.

Keywords

detekcija anomalij;onesnaženost zraka;strojno učenje;XGBoost;LSTM samokodirnik;matrični profil;časovne vrste;visokošolski strokovni študij;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2024
Typology:	2.11 - Undergraduate Thesis
Organization:	UL FRI - Faculty of Computer and Information Science
Publisher:	[L. Volk]
UDC:	004.85:502.3:613.15(043.2)
COBISS:	211114755
Views:	130
Downloads:	39
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Anomaly detection in air pollutant measurements
Secondary abstract:	Air pollution is a serious environmental problem that can negatively impact human health and environmental quality. This thesis focuses on anomaly detection in air pollutant measurements: sulfur dioxide (SO2), ozone (O3), nitrogen dioxide (NO2), nitrogen oxide (NO), carbon monoxide (CO), and particulate matter PM10. The data was obtained from ARSO. Detecting anomalies in these measurements is crucial for ensuring reliable data, as anomalies can indicate sensor malfunctions or other exceptional events. In this thesis, we implemented and compared three different machine learning models: XGBoost, LSTM autoencoder, and matrix profile. Among the analyzed methods, the XGBoost model performed the best, successfully detecting the highest number of anomalies and achieving the highest evaluation metrics.
Secondary keywords:	anomaly detection;air pollution;machine learning;XGBoost;LSTM autoencoder;matrix profile;time series;computer science;diploma;Strojno učenje;Onesnaževalci zraka;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS):	Bachelor thesis/paper
Study programme:	1000470
Embargo end date (OpenAIRE):	1970-01-01
Thesis comment:	Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages:	1 spletni vir (1 datoteka PDF (40 str.))
ID:	24920789