Short-Term Load Forecasting Based on Big Data Technologies

With the construction of smart grid, lots of renewable energy resources such as wind and solar are deployed in power system. It might make the power system load varied complex than before which will bring difficulties in short-term load forecasting area. To overcome this issue, this paper proposes a new short-term load forecasting framework based on big data technologies. First, a cluster analysis is performed to classify daily load patterns for individual loads using smart meter data. Next, an association analysis is used to determine critical influential factors. This is followed by the application of a decision tree to establish classification rules. Then, appropriate forecasting models are chosen for different load patterns. Finally, the forecasted total system load is obtained through an aggregation of an individual load's forecasting results. Case studies using real load data show that the proposed new framework can guarantee the accuracy of short-term load forecasting within required limits.