Home // ENERGY 2022, The Twelfth International Conference on Smart Grids, Green Communications and IT Energy-aware Technologies // View article

Improved Data Preprocessing Approach to Short-Term Load Forecasting

Authors:
Athanasios Ioannis Arvanitidis
Dimitrios Bargiotas

Keywords: short-term load forecasting, data preprocessing, scaling techniques, multi-layer perceptron

Abstract:
One of the most critical aspects for the smooth operation of power systems is short-term load forecasting. Forecast accuracy has a significant impact on an electricity utility's economic viability and reliability. Thus, robust deep learning methods, such as artificial neural networks, are implemented in order to achieve higher accuracy load forecasting results. In this paper, a new preprocessing method of the input data of a neural network, which emphasizes on the importance of specific input data, that show a higher Pearson’s correlation coefficient with the output result, is proposed. This work implements the proposed preprocessing technique and compares the results with those derived from the classical min-max scaling methods. Numerical results of next hour’s load forecasting, based on a multi-layer perceptron with the implementation of the proposed data scaling approach, show higher precision than the typical scaling method, demonstrating the importance of our work.

Pages: 6 to 10

Copyright: Copyright (c) IARIA, 2022

Publication date: May 22, 2022

Published in: conference

ISSN: 2308-412X

ISBN: 978-1-61208-967-6

Location: Venice, Italy

Dates: from May 22, 2022 to May 26, 2022