Smart Electricity Meter Load Prediction in Dubai Using MLR, ANN, RF, and ARIMA
Open Access
- 12 January 2023
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 12 (2) , 389
Abstract
Load forecasting is one of the main concerns for power utility companies. It plays a significant role in planning decisions, scheduling, operations, pricing, customer satisfaction, and system security. This helps smart utility companies deliver services more efficiently and analyze their operations in a way that can help optimize performance. In this paper, we propose a study of different techniques: multiple linear regression (MLR), random forests (RF), artificial neural networks (ANNs), and automatic regression integrated moving average (ARIMA). This study used electricity consumption data from Dubai. The main objective was to determine the load demand for the next month in the whole country and different municipal areas in Dubai, as well as to assist a utility company in future system scaling by adding new power stations for high-demand regions. The results showed that the accuracy of using ARIMA was about 93% when working with only a single district, but both ANN and RF achieved excellent accuracy of about 97% in all cases. In addition, the mean absolute percentage errors improved from 2.77 and 2.17 to 0.31 and 0.157 for ANN and RF, respectively, after anomaly elimination and the use of our proposal. Therefore, the use of an ANN for such data types is recommended in most cases, particularly when working on a complete dataset. Additionally, both the ANN and RF models are good choices when working on a single-category region because they both attained the same accuracy of almost 91.02 percent.Keywords
This publication has 49 references indexed in Scilit:
- Data Preparation2017
- An Integrated IoT Architecture for Smart MeteringIEEE Communications Magazine, 2016
- The anomalous data identification study of reactive power optimization system based on big data2016
- Data Mining with SPSS Modeler2016
- Generalized Principal Component Analysis2016
- Short-term load forecasting based on big data technologiesCSEE Journal of Power and Energy Systems, 2015
- Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literatureGeoscientific Model Development, 2014
- A Study on Multiple Linear Regression AnalysisProcedia - Social and Behavioral Sciences, 2013
- ARIMA models to predict next-day electricity pricesIEEE Transactions on Power Systems, 2003
- Random ForestsMachine Learning, 2001