Big Data Analytics in a Day-Ahead Electricity Price Forecasting Using TensorFlow in Restructured Power Systems

Although Smart Grids (in power systems) can bring beneficial features such as lowering the power market price, it enforces having a more comprehensive platform that can analyze a huge amount of data coming from sensor networks throughout the system. In this paper, a distributed and highly scalable machine learning platform called Tensorflow has been studied. Tensorflow is a Google's open source platform for artificial intelligence systems. The advantage of this package is that it can run parallel computations on distributed systems on a scale of thousands cores. Hence, big data analytics for real time applications are nowadays done with this package. In this paper, a case study considering a day-ahead price forecasting for power market of two European countries have been studied via Tensorflow. All the data gathered from the renewable energies and conventional generators, as well as weather forecasts, have been fed to the forecasting system to predict the day-ahead price of the market. The result shows improvement in the convergence speed as well as the prediction accuracy.