Research
Price forecasting and operational decisions in the National Electricity Market.
We research deep-learning methods for wholesale electricity price forecasting in Australia's National Electricity Market, and the downstream operational decisions those forecasts unlock, above all battery storage dispatch. Methods are evaluated on held-out AEMO market periods against the official predispatch baseline, with particular attention to extreme-price events where forecast value concentrates.
Projects
Active research threads. Each project carries its own paper trail and, where it makes sense, a case study or live demo.
- activeforecastingdeep learning
NEM electricity price forecasting
Forecasting wholesale electricity prices in the National Electricity Market under extreme volatility and renewable-driven structural shifts — multi-horizon and multi-region forecasting, spike-aware modelling, and regime-adaptive methods targeting the conditions where forecast value concentrates.
- in progressBESSRLMARL
AI-Driven Battery Orchestration
We develop AI-driven coordination strategies for household and community batteries, enabling distributed assets to operate as an intelligent energy network. Using multi-agent reinforcement learning, the project balances household value with community objectives such as peak reduction, renewable utilisation, market participation and battery longevity.
Case study in preparation.
Publications
2026
- preprint↗ forecasting
Mohammed Osman Gani, Zhipeng He, Chun Ouyang, Sara Khalifa
Deep Time-Series Models Meet Volatility: Multi-Horizon Electricity Price Forecasting in the Australian National Electricity Market
arXiv 2602.01157
Accurate electricity price forecasting is increasingly difficult in markets characterised by extreme volatility, frequent price spikes, and rapid structural shifts. Deep learning has been increasingly adopted in EPF due to its ability to achieve high forecasting accuracy.
Publication list is being finalized — contact us for preprints.
People
Chun Ouyang
Profile ↗Energy Transition Centre & School of Information Systems, QUT
Sara Khalifa
Profile ↗Energy Transition Centre & School of Information Systems, QUT
Zhipeng He
Profile ↗Energy Transition Centre & School of Information Systems, QUT
Mohammed Osman Gani
Profile ↗PhD Candidate↗ forecastingEnergy Transition Centre & School of Information Systems, QUT
Adalie Hoang
MPhil Student↗ bessSchool of Information Systems, QUT
Affiliation & collaborate
A research project of Queensland University of Technology (QUT). Market data via AEMO.
Code and datasets are released alongside publications. We welcome industry pilots, student supervision, and academic collaboration.
Get in touch