Aplikace strojového učení při předpovědi výskytu mlh na území ČR = Application of machine learning in fog prediction in the Czech Republic

To chci

Ground fog presents a potential danger, especially in traffic, and its early warning can save not only money but also human lives. The prediction of the fog occurrence is currently based on a combination of outputs from numerical weather prediction models and a subjective evaluation of the situation by an experienced meteorologist. The aim of this contribution is to present a new approach using available meteorological data in the fog occurrence prediction based on machine learning methods. For this purpose, we use the classification algorithm of the decision tree ensemble, which is logically close to the human way of thinking. Station measurements and numerical predictions from the ALADIN/CZ model for the period 2016 to 2018 were used to learn the model. The evaluation of results for 2019 showed that the machine learning model provides more accurate point prediction of fog up to 36 hours compared to the raw numerical model. The proposed machine learning model can be easily applied to the operation of meteorological forecasting, where it can be used not only as a basis for creating forecasts, but also as an objective tool for identifying the causes of fog. … celý popis

Podrobná bibliografie

Hlavní autor

Patrik Benáček

Další autoři

Pavel Borovička, 1981-

Typ dokumentu

Články

Fyzický popis

Ilustrace

Publikováno v

Meteorologické zprávy. -- ISSN 0026-1173. -- Roč. 74, č. 5 (2021), s. 141-148

Témata

CZECH REPUBLIC

WEATHER FORECASTING

FOG

DATA COLLECTION

CLASSIFICATION

DATA PROCESSING

METEOROLOGICAL STATIONS

HABITATS

TIME

FORESTS

MODELS

Artificial intelligence

EVALUATION

přízemní mlhy

algoritmus náhodného lesa

strojové učení

Česká republika

předpověď počasí

mlha

sběr dat

klasifikace

zpracování dat

meteorologické stanice

stanoviště

čas

lesy

modely

umělá inteligence

hodnocení

více… méně…

Popis jednotky

1 obrázek, 5 grafů, 3 mapy, 2 schémata, 1 tabulka

Bibliografie

Literatura na s. 148