Forecasting electricity demand of Bangladesh and its relation with physical and natural variables using group method of data handling model

Abstract

The purpose of this work is to forecast the electricity demand of Bangladesh using the Group Method of Data Handling (GMDH). Electricity is one of the key variables in ensuring economic growth and a higher standard of life. For a developing country like Bangladesh, it is very important to have a near accurate forecast of electricity demand for future planning. After careful preparation of a time series data set of GDP, GNI, CO2 emission, ambient temperature, the data sheet was fed into the GMDH model and an acceptable forecast was found. GMDH Shell software is used for the application of GMDH algorithm and time series analysis.This purpose is also resolved by deep learning and ANN application.High extensive correlation among features and pre processing could make fruitful in optimal prediction