master's thesis
Lojze Žust (Author), Matej Kristan (Mentor), Matjaž Ličer (Co-mentor)

Abstract

Natančne kratkoročne napovedi višine morske gladine so ključne za pravočasno detekcijo ekstremnih dogodkov, zagotavljanje varnosti prebivalstva in omejitve povzročene ekonomske škode. Astronomski in vremenski vplivi predstavljajo glavni del sprememb višine morske gladine. V delu predlagamo metodo HIDRA, ki predstavlja nov, residualen pristop za napovedovanje višine gladine -- z ločenim modeliranjem in odstranitvijo astronomskega vpliva iz plimnega signala, ločimo posamezna vpliva ter zgradimo mrežo v celoti posvečeno modeliranju kompleksnejšega vremenskega vpliva na spremembo višine gladine. HIDRA uvaja učljiv prostorski vremenski kodirnik ter fuzijo informacij vremenskega in plimnega vpliva v celovito mrežo, ki omogoča pripravo diskriminativnih značilk za problem napovedovanja višine morske gladine. Analiza na dveh ločenih podatkovnih zbirkah za napovedovanje višine gladine (Koper in Acqua Alta) kaže visoko generalizacijsko sposobnost predlagane metode. V primerjavi s trenutno najboljšim numeričnim modelom NEMO, HIDRA doseže 38% manjši RMSE v splošnem in 41% manjši RMSE na poplavnih dogodkih. Hkrati ima HIDRA mnogo manjšo računsko kompleksnost ter skrajša procesorski čas izvajanja metode za faktor več kot pol milijona, na manj kot sekundo in posledično bistveno zmanjša energijski okoljski odtis napovedovanja višine morske gladine.

Keywords

sea level;forecasting;tide;storm surges;deep learning;computer science;computer and information science;master's thesis;

Data

Language: English
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [L. Žust]
UDC: 004.85:627.512(043.2)
COBISS: 31839235 Link will open in a new window
Views: 30
Downloads: 8
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: Slovenian
Secondary title: A deep learning method for storm surge forecasting
Secondary abstract: Accurate short-term sea level forecasting is essential for early detection of extreme sea level events such as storm surges in order to ensure public safety and reduce the impact on coastal economies. Sea level is mainly influenced by astronomic and atmospheric factors. We propose HIDRA, a novel residual approach to sea level forecasting -- by estimating and subtracting the astronomic influence from the tidal signal using standard physics-based approaches, we disentangle the two influences and build a network to fully focus on the more complex atmospheric-based part of sea level fluctuations. HIDRA introduces a trainable atmospheric spatial encoder and feature fusion of atmospheric and tidal features into an end-to-end network, which enables discriminative feature construction for the task of sea level prediction. Evaluation on two sea level forecasting tasks (Koper and Acqua Alta) demonstrates a great generalization capability of HIDRA. In comparison with the state-of-the-art numerical NEMO model, HIDRA achieves 38% lower RMSE in general and 41% lower RMSE on storm surge events, while having vastly lower computational complexity -- HIDRA achieves more than half a million times lower CPU times, producing predictions in a fraction of a second and thus significantly reducing the energy footprint of sea level prediction.
Secondary keywords: višina gladine;napovedovanje;plima;poplavljanje;globoko učenje;računalništvo;računalništvo in informatika;magisteriji;
Type (COBISS): Master's thesis/paper
Study programme: 1000471
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: XVI, 73 str.
ID: 12692813