z-logo
open-access-imgOpen Access
Design of meteorological information database for forecasting and clustering in microgrid system
Author(s) -
Volodymyr Osypenko,
Nikolay Kiktev,
T. Lendiel
Publication year - 2020
Publication title -
energetika ì avtomatika
Language(s) - English
Resource type - Journals
ISSN - 2223-0858
DOI - 10.31548/energiya2020.04.055
Subject(s) - computer science , database , cluster analysis , sql , relational database , microgrid , data mining , control (management) , artificial intelligence
To build Microgrid systems, it is necessary to obtain data from the meteorological service, process them and make decisions about which source of electricity is advisable to use at a given time of day, season, under current weather conditions. The aim of the study is to develop and create a distributed information system database for cluster analysis, processing and storage of incoming meteorological data, a weather forecasting algorithm based on the values of the selected indicators to further determine the type of alternative energy sources used based on the forecast. The article describes designed and implemented distributed information system for reading from the Internet, storing and further processing meteorological data for any region with the aim of forecasting for the effective use of renewable energy sources in Microgrid system. The project is implemented on the basis of a relational database Microsoft SQL Server. Each of the tables has fields that describe the weather conditions necessary to solve the task – to determine the source of electricity, the use of which is cost-effective in a given period of the year, time of day, geographical location and weather conditions. The application that operates with a database has been developed in C # according to the Windows Forms Application template. The distribution of temperature indicators is realized depending on the time of the conducted research for a certain period using cluster analysis. Forecasting weather data is performed using an autoregressive time series model. The user interface was created with Microsoft Visual Studio tools. All data processing is performed on the local server side.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here