View in EDS

Wind speed monitoring system for GFS and ECMWF data using AWS Grafana.

Saved in:
Bibliographic Details
Title: Wind speed monitoring system for GFS and ECMWF data using AWS Grafana.
Authors: Asmawi, Tengku Nazmi Tengku1 (AUTHOR) tgnazmi98@gmail.com, Kassim, Murizah1,2 (AUTHOR) murizah@uitm.edu.my, Jumaat, Abdul Kadir2,3 (AUTHOR) abdulkadir@tmsk.uitm.edu.my, Zain, Jasni Mohamad2,3 (AUTHOR) jasni67@uitm.edu.my, Haron, Nazleeni Samiha4 (AUTHOR) nazleeni@utp.edu.my, Jaafar, Jafreezal4 (AUTHOR) jafreez@utp.edu.my, Ibrahim, Siti Sara1,5 (AUTHOR) saraibrahim@ns.uitm.edu.my, Yusoff, Marina2,3 (AUTHOR) marina998@uitm.edu.my, Tahir, Nooritawati Md1,6 (AUTHOR) noori425@uitm.edu.my, Mausor, Farahida Hanim4 (AUTHOR) farahida_17010728@utp.edu.my, Krishnan, Nor Farisha Muhamad4 (AUTHOR) farisha.krishnan@utp.edu.my
Source: AIP Conference Proceedings. 2025, Vol. 3322 Issue 1, p1-11. 11p.
Subject Terms: *WIND speed measurement, *WEATHER forecasting, *DATA visualization, *RENEWABLE energy sources, *LONG-range weather forecasting, *METEOROLOGICAL databases, *DATA visualization software, *RECURRENT neural networks
Company/Entity: EUROPEAN Centre for Medium-Range Weather Forecasts (Organization) , AMAZON Web Services Inc.
Abstract: Effective wind speed monitoring is critical for industries such as renewable energy and weather forecasting. This paper introduces an advanced Wind Speed Monitoring System that combines AWS Grafana with data from the Global Forecast System (GFS) and the European Centre for Medium-Range Weather Forecasts (ECMWF), integrated with a Long Short-Term Memory (LSTM) prediction model for enhanced forecasting accuracy. The system leverages high-resolution wind speed data from GFS and ECMWF, renowned for their reliability and extensive global coverage. This data is ingested into Amazon Web Services (AWS) for processing and storage using Amazon Timestream, a time-series database optimized for handling large volumes of data efficiently. To improve the predictive capabilities of the system, an LSTM neural network model is employed. LSTM, a type of recurrent neural network (RNN), excels at capturing temporal dependencies and patterns in time-series data, making it ideal for predicting wind speed. The LSTM model is trained using historical wind speed data from GFS and ECMWF, allowing it to generate accurate short-term and long-term wind speed forecasts. A Visualization Dashboard (VD) using AWS Grafana was developed with dynamic and interactive real live and LSTM prediction of wind speed data. The main shows a wind speed monitoring system home page. Other pages are the tropical depression of wind speed, parameters data of wind speed, the model performance comparison page of the monitoring system and model comparison page. The Grafana VD shows robust querying and visualization of trends analysis, compare forecasts from GFS, ECMWF, and the LSTM model, and set up alerts for significant changes in wind speed. This Wind Speed Monitoring System is significant in providing a comprehensive of accurate wind speed forecasting and monitoring. The system's architecture, implementation, and benefits are discussed, highlighting its potential to enhance decision-making processes across various industries through improved wind speed data analysis and visualization. [ABSTRACT FROM AUTHOR]
Database: Academic Search Index
Description
Abstract:Effective wind speed monitoring is critical for industries such as renewable energy and weather forecasting. This paper introduces an advanced Wind Speed Monitoring System that combines AWS Grafana with data from the Global Forecast System (GFS) and the European Centre for Medium-Range Weather Forecasts (ECMWF), integrated with a Long Short-Term Memory (LSTM) prediction model for enhanced forecasting accuracy. The system leverages high-resolution wind speed data from GFS and ECMWF, renowned for their reliability and extensive global coverage. This data is ingested into Amazon Web Services (AWS) for processing and storage using Amazon Timestream, a time-series database optimized for handling large volumes of data efficiently. To improve the predictive capabilities of the system, an LSTM neural network model is employed. LSTM, a type of recurrent neural network (RNN), excels at capturing temporal dependencies and patterns in time-series data, making it ideal for predicting wind speed. The LSTM model is trained using historical wind speed data from GFS and ECMWF, allowing it to generate accurate short-term and long-term wind speed forecasts. A Visualization Dashboard (VD) using AWS Grafana was developed with dynamic and interactive real live and LSTM prediction of wind speed data. The main shows a wind speed monitoring system home page. Other pages are the tropical depression of wind speed, parameters data of wind speed, the model performance comparison page of the monitoring system and model comparison page. The Grafana VD shows robust querying and visualization of trends analysis, compare forecasts from GFS, ECMWF, and the LSTM model, and set up alerts for significant changes in wind speed. This Wind Speed Monitoring System is significant in providing a comprehensive of accurate wind speed forecasting and monitoring. The system's architecture, implementation, and benefits are discussed, highlighting its potential to enhance decision-making processes across various industries through improved wind speed data analysis and visualization. [ABSTRACT FROM AUTHOR]
ISSN:0094243X
DOI:10.1063/5.0290599