Zone Wise Hourly Load Prediction Using Regression Decision Tree Model

This paper presents a predictive model to compute regional power demand on an hourly interval using regression decision tree machine learning algorithm. The training dataset has been derived from a zone wise hourly load supplied record of power grid company of Bangladesh limited (PGCB) for three reg...

Full description

Saved in:

Bibliographic Details
Published in:	2018 International Conference on Innovation in Engineering and Technology (ICIET) pp. 1 - 6
Main Authors:	Chowdhury, Dhiman, Sarkar, Mrinmoy, Haider, Mohammad Zakaria, Alam, Taufique
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-12-2018
Subjects:	Hourly load load forecasting Load modeling machine learning Machine learning algorithms Mathematical model power demand prediction Prediction algorithms Predictive models prune regression decision tree Regression tree analysis
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	This paper presents a predictive model to compute regional power demand on an hourly interval using regression decision tree machine learning algorithm. The training dataset has been derived from a zone wise hourly load supplied record of power grid company of Bangladesh limited (PGCB) for three regions-Dhaka, Chittagong and Rajshahi for six consecutive days in March, 2018. A regression decision tree based predictive model has been developed for substantial load forecasting application. Computing power data time instants ahead from a historical dataset conforms to an efficient tool for maintaining a balance between demand and supply along with for making preparedness in case of a contingency. The regression algorithm yields to without and with pruning applications to evaluate performance. The predictive model has been simulated on MATLAB and the performance assessments corroborate the reliability of the framework.
DOI:	10.1109/CIET.2018.8660781