Machine Learning Approaches to Electricity Consumption Forecasting in Automated Metering Infrastructure (AMI) Systems: An Empirical Study

In a Smart grid, implementation of value-added services such as distribution automation (DA) and Demand Response (DR) [1] rely heavily on the availability of accurate electricity consumption forecasts. Machine learning based forecasting systems, due to their ability to handle nonlinear patterns, appear promising for the purpose. An empirical evaluation of eight machine learning based systems for electricity consumption forecasting, based on Extreme Learning machines (ELM), Ensemble Regression Trees (ERT), Artificial Neural Network (ANNs) and regression is presented in this study. Forecasting systems thus designed, are validated on consumption data collected from 5275 users. Result indicate that ELM based electricity consumption forecasting systems are not only more accurate than other systems considered, they are considerably faster as well.

[1]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[2]  Rob J Hyndman,et al.  25 years of time series forecasting , 2006 .

[3]  Binoy B. Nair,et al.  A Stock Trading Recommender System Based on Temporal Association Rule Mining , 2015 .

[4]  Binoy B. Nair,et al.  Modeling of Consumption Data for Forecasting in Automated Metering Infrastructure (AMI) Systems , 2016, CSOC.

[5]  Ram Rajagopal,et al.  Household Energy Consumption Segmentation Using Hourly Data , 2014, IEEE Transactions on Smart Grid.

[6]  Donald F. Specht,et al.  The general regression neural network - Rediscovered , 1993, Neural Networks.

[7]  Pierluigi Siano,et al.  Demand response and smart grids—A survey , 2014 .

[8]  Tao Hong,et al.  Long Term Probabilistic Load Forecasting and Normalization With Hourly Information , 2014, IEEE Transactions on Smart Grid.

[9]  Guang-Bin Huang,et al.  An Insight into Extreme Learning Machines: Random Neurons, Random Features and Kernels , 2014, Cognitive Computation.

[10]  Binoy B. Nair,et al.  Artificial intelligence applications in financial forecasting - a survey and some empirical results , 2015, Intell. Decis. Technol..

[11]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[12]  Zhaohui Tang,et al.  The review of demand side management and load forecasting in smart grid , 2016, 2016 12th World Congress on Intelligent Control and Automation (WCICA).

[13]  Hassan Ghasemi,et al.  Residential Microgrid Scheduling Based on Smart Meters Data and Temperature Dependent Thermal Load Modeling , 2014, IEEE Transactions on Smart Grid.

[14]  N. R. Sakthivel,et al.  Clustering stock price time series data to generate stock trading recommendations: An empirical study , 2017, Expert Syst. Appl..

[15]  Binoy B. Nair,et al.  An intelligent recommender system for stock trading , 2015, Intell. Decis. Technol..

[16]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[17]  Yik-Chung Wu,et al.  Load/Price Forecasting and Managing Demand Response for Smart Grids: Methodologies and Challenges , 2012, IEEE Signal Processing Magazine.

[18]  Guang-Bin Huang,et al.  Extreme learning machine: a new learning scheme of feedforward neural networks , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).