Long-Term Electricity Load Forecasting Based On Cascade Forward Backpropagation Neural Network

Authors

  • Widi Aribowo State University of Surabaya, Indonesia
  • Supari Muslim State University of Surabaya, Indonesia

Keywords:

CFBNN, Long-Term Load Forecasting, MAPE, Neural Network,

Abstract

Nowadays, the Electrical System has an important role in all sectors of life. Electricity has a strategic role. Accuracy and reliability in electricity load forecasting is a great key that can help electricity companies in supplying electricity efficiency, hence, reducing wasted energy. In addition, electricity load forecasting can also help electricity companies to determine the purchase price and power generation. Long-term forecasting is a method of forecasting with a span of more than one year. The historical data will be a reference in solving the problems. This research propose the concept of cascade forward backpropagation for long-term load forecasting. The advantage of this concept is that it can accommodate non-linear conditions without ignoring the linear conditions. This study compared the results of the original data, Feed Forward Backpropagation Neural Network (FFBNN) and Cascade Forward Backpropagation Neural Network (CFBNN). The results were measured by comparing Mean Absolute Deviation (MAD) and Mean Absolute Percentage Error (MAPE).

Downloads

Download data is not yet available.

Downloads

Published

2020-06-30

How to Cite

Aribowo, W., & Muslim, S. (2020). Long-Term Electricity Load Forecasting Based On Cascade Forward Backpropagation Neural Network. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 12(2), 39–44. Retrieved from https://jtec.utem.edu.my/jtec/article/view/5644

Issue

Section

Articles