1 School of Electrical and Electronic Engineering, SungKyunKwan University, Republic of Korea
2 School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore
3 Department of Electrical Engineering, Hong Kong Polytechnic University, Hong Kong
Artificial Neural Network(ANN) is an effective approach for short-term load forecasting (STLF). While accurate load forecasting is pivotal for the economic and secure operation of the power system, there have been continuous efforts to achieve high load forecasting accuracy. There are two main ways to do this: the first is the improvement of the performance of the learning algorithm, and the second is how well the data features are extracted through the pre-processing process. In this paper, we design Hierarchical-Extreme Learning Machine (H-ELM) based model for forecasting the electricity load of Australian National Electricity Market (NEM) data. Owing to the very fast training/tuning speed of feed-forward neural network and multilayer concept, the H-ELM model, like the Extreme Learning Machine (ELM), is able to make fast and efficient predictions, while overcoming the instability of the forecast, which was a drawback of ELM. In addition, the H-ELM model shows better performance due to data pre-processing.
- input
- ouput
[1] Short-term load forecasting of Australian National Electricity Market by an ensemble model of extreme learning machine
[2] Extreme Learning Machine for Multilayer Perceptron, IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, VOL. 27, NO. 4, APRIL 2016