FAQs

Languages
Vietnamese
English
Home
About VinIF
VINIF.2023.DA019 – Machine Learning for Typhoon Formation Prediction
Home
About VinIF
VINIF.2023.DA019 – Machine Learning for Typhoon Formation Prediction

VINIF.2023.DA019 – Machine Learning for Typhoon Formation Prediction

project manager image
Principle Investigator
Assoc.Prof. Nguyen Thi Nhat Thanh
Host Organization
University of Technology and Engineering, Vietnam National University

Storms are one of the main natural causes of property and life damage in Vietnam. In applied research on storm forecasting, an important question is how storm forecasting can be improved for risk management and disaster prevention. Seeking to understand more about the mechanics of storm formation, we recently reviewed and analyzed in detail various physical aspects of the storm formation process, using theoretical models and simulations. The results of our initial analysis suggest that there are physical constraints in the tropics that could potentially help improve forecasts of hurricane formation. These preliminary results are the starting point for this research topic. Besides, with some very positive initial results of our recent testing of machine learning methods for storm formation forecasting, the research project includes three main goals:

  • Develop a number of machine learning and deep learning methods to forecast storm formation in the East Sea and northwest Pacific regions from climate data and satellite images.
  • Applying new machine learning methods to identify climatic factors that govern storm formation near Vietnam’s coastal areas to help predict and respond to future climate change in Vietnam.
  • Strengthen research capacity and promote multidisciplinary collaboration in the fields of climate research, weather forecasting, and machine learning applications, to provide practical solutions to enhance the quality of storm forecasts in Vietnam.
project manager image
Principle Investigator
Assoc.Prof. Nguyen Thi Nhat Thanh
Host Organization
University of Technology and Engineering, Vietnam National University

Tags

Expect Progress
01/11/2023
31/10/2024
Phase 1

– Collecting and preprocessing satellite, meteorological, and weather data for typhoon formation prediction in the South China Sea and northwest Pacific Ocean regions;
– Developing an Application Programming Interface (API) for data preprocessing to support testing, evaluation, and validation of machine learning algorithms;
– Developing and evaluating the effectiveness of machine learning models for typhoon formation prediction using satellite and meteorological data;
– Submitting and presenting a report at the first international workshop/conference;
– Drafting the first Q1 paper on the application of machine learning methods for typhoon formation prediction and evaluating the effectiveness of the method with different input datasets;
– Writing the Year 1 report;
– Collecting and preprocessing data for future weather scenario simulations;

31/10/2025
Phase 2

– Evaluate the sensitivity of machine learning models with different climate and/or image input data to provide well-calibrated probability predictions for models developed in Year 1.
– Assess the convergence and effectiveness of selected machine learning methods with different features based on identifying factors on a large scale that influence typhoon formation in the coastal areas of Vietnam.
– Analyze and optimize machine learning models to achieve the best results in predicting typhoon formation at each future time point for the coastal areas of Vietnam.
– Revise and submit the draft of the first Q1 paper.
– Submit and present a report at the second international workshop/conference.
– Draft a second Q1 paper on the performance of machine learning models.
– Draft a paper for a national journal on the application of machine learning models for typhoon formation prediction.
– Draft a third Q1 paper on the design and testing of typhoon formation prediction using machine learning and evaluate the effectiveness of this method compared to other forecasting methods.
– Write the Year 2 report.

31/10/2026
Phase 3

– Implement machine learning algorithms developed during Years 1-2 to forecast typhoon formation at the National Center for Hydro-Meteorological Forecasting.
– Quantitatively evaluate the effectiveness of the proposed machine learning model in typhoon formation prediction compared to the forecasting models currently used in Vietnam.
– Submit the draft of the second Q1 paper.
– Submit the draft of the third Q1 paper.
– Submit the draft of the paper for the national journal starting from Year 2.
– Submit and present a report at the third international workshop/conference.
– Write a technical report for the development of a computer program for typhoon formation prediction using machine learning methods at the National Center for Hydro-Meteorological Forecasting.
– Write a project synthesis report.

Tags

VINGROUP INNOVATION FOUNDATION

Address: Floor 9, Century Tower, Times City, 458 Minh Khai Street, Hanoi, Vietnam

SPONSORSHIP PROGRAM

Follow Us

Twitter Facebook Linkedin Youtube

Copyright © 2020. All Rights Reserved.

Visit counter 463561

Twitter Facebook Linkedin Youtube