Physics informed machine learning for energy forecasting

Overview

https://research.qut.edu.au/energytransition/https://research.qut.edu.au/energytransition/

Accurate forecasting is at the heart of many modern industries from energy and transport to retail, supply chains, finance, climate, and health. This research project explores deep learning approaches for time-series forecasting, investigating how modern architectures such as recurrent neural networks, LSTMs, Temporal Convolutional Networks, transformers, and multimodal foundation models can shape the next generation of forecasting systems.

The overarching goal is to develop robust, interpretable, and scalable forecasting models that outperform classical methods and work effectively in real-world settings, including domains with partial information, high uncertainty, and rapidly changing environments.

This project suits students who enjoy machine learning, time series, programming, and applied research with potential impact across multiple industries.

Research activities

You will be part of the Forecasting Insights and Decision Making Group at QUT. Students will work on a structured set of research activities, tailoring their focus to sectoral interests (e.g. energy, retail, transport, climate, logistics):

• survey deep learning models for forecasting
• identify gaps in existing research
• work with real or simulated datasets (energy demand/solar, retail sales, traffic flows, commodity prices, etc)
• perform time-series cleaning, scaling, feature engineering, and handling missing data
• build pipelines for training/validation/testing and rolling-origin evaluation.

Outcomes

The expected outcomes are:

• a deep learning forecasting model
  • designed, trained, and evaluated on real-world or simulated data
• a technical research report
  • documenting methods, experiments, and insights
• a publishable paper
  • a short manuscript suitable for submission to a forecasting or applied ML venue.

Skills and experience

• Deep learning.

Keywords

• forecasting
• deep learning

Contact

Contact the supervisor via email mahdi.abolghasemi@qut.edu.au.