Loss of data management in real-time short-term forecasting algorithms

Abstract

In this paper, a new technique is proposed to mitigate the problem of data loss in real-time short-term forecasting (STF) algorithms. The proposed method can be applied to any STF algorithm in order to resolve the problem of data loss with minimal additional computation. In addition, the proposed implementation strategy can be employed to address the challenges of handling holidays in the electrical load and energy price forecasting algorithms. In order to evaluate the performance of the proposed method, a well-accepted real-time short-term load forecasting (STLF) method has been implemented. The historical electricity load information and ambient temperature of a real-life load are used in this study. The investigation reveals that the proposed method properly addresses the above-mentioned challenges with high accuracy and minimal additional computation. Although the case study in this paper deals with the practical challenges only in real-time STLF, the proposed method can be used in any real-time STF algorithm, such as renewable power generation and energy price forecasting.