Cloud Computing for Short-Term Load Forecasting Based on Machine Learning Technique

摘要

Short-term electric load forecasting (STLF) plays the main role in making operational decisions in any electrical power system. The implementation of forecasting algorithms collides with the high computational power needed to perform the complex perdition processes on large datasets. In this paper, a cloud-based STLF algorithm is implemented. The performance analysis of the proposed system was compared against the implementation of the same algorithm on a local machine and against many other forecasting algorithms. The results show that the cloud-based implementation enhances the algorithm execution time.