Renewable Energy Prediction/Forecasting

School of Engineering and Technology
Biplob Ray
Narottam Das


The increasing demand for and growing dependency on Renewable Energy System (RES) necessitates the need for renewable power systems that are less volatile and more predictable. Due to the variability in renewable energy, it is important to have both a short and a long term, accurate forecast of available solar and wind energy for power system planning studies. This forecasting can be used for day to day operations or for strategic management decisions. Furthermore, metadata collected by the Internet of Things (IoT) systems can be used for fault prediction and health monitoring of renewable energy stations and rechargeable batteries. In this project, you will be using historic solar and wind data for renewable power forecasting in the power system planning horizon. You will also have scope to design an IoT system to collect live sensors data of Renewable Energy Systems and storage device for fault prediction and health status monitoring for remote maintenance.

  • Review of the related literature to define the best methodology to approach and frame the problem.
  • Applying machine learning algorithms to forecast short- and long-term PV and wind energy based on historic and non-historic data and comparing the results with those of the state-of-the-art approaches in this domain.
  • Assess the probabilistic reserve required for the Australian interconnected system with high penetration of solar and wind generations.
  • Predict fault for a secure renewable power source
  • Create health monitoring for renewable energy system and storage facilities.
Environmental Sciences| Information and Computing Sciences| Engineering
Machine learning, Smart grid, Renewable forecasting
October, 2020
Melbourne| Online| Offshore| By Negotiation

Project contacts