An Intelligent System for Forecasting the Trend of Consumed Electricity

In big data era, more and more people concern on what hidden knowledge can be found from data. Today, big data is not only the data scenario with large volume, but also high-speed and changing all the time. Such data streams commonly exist in Smart Grid facilities. As previous research, incremental learning method was proposed to discover the decision model from the continuous data streams. The decision model is able to interpret the findings to an easily understood format that can be used by humans and machines. In this paper, we investigate the previous theories of incremental learning, and apply them for constructing a streaming process engine in power grid system. The advanced learning method produces an efficient way to handle the high-speed data streams that are captured from power grid units, and establishes a decision support system to forecast the trend of power load in certain period.