TRAP@NCI

Wind Electricity Generation Forecasting using Time Series Analysis Techniques in India

Gupta, Sonali (2018) Wind Electricity Generation Forecasting using Time Series Analysis Techniques in India. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (2MB) | Preview

Abstract

Electricity crisis is a major concern for developing countries and this problem can be fulfilled by various technologies of renewable energy such as water energy, solar energy, wind energy and many more. The project presented the idea of forecasting electricity generation using wind speed in two different cities of India: Jodhpur and Bengaluru. This thesis investigates the use of time series models for wind electricity forecast for both the cities. A contrast for four time series models i.e. TBATS, ARIMA, SES, and HOLTS is conducted and results are presented.The result of this study shows that the TBATS model is outperformed all the other developed models. The key finding is that TBATS model able to deliver accurate forecast of wind speed with MASE of 0.76%, RMSE of 1.41 %, MAE of 1.19%, and MAPE of 0.34% for 8 months forecast horizon. The results of forecasted value of best performing TBATS model are used in the computation of electricity generation for both cities finalized that on the same investment for setting up wind power plant, electricity produced by Jodhpur is more than the Bengaluru city. Hence, Jodhpur is a better location to set up the wind power plant.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electricity Supply
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: CAOIMHE NI MHAICIN
Date Deposited: 05 Nov 2018 10:39
Last Modified: 05 Nov 2018 10:39
URI: http://trap.ncirl.ie/id/eprint/3424

Actions (login required)

View Item View Item