JapanEnergy and utilities2019 – ongoingChugoku Electric Power
AI inflow forecasting and hydropower generation planning
Chugoku Electric's hydropower trial reduced annual dam spill by about 6 million cubic metres
Problem
Rainfall forecasts did not reliably predict the amount and timing of water entering dams. Operators struggled to schedule generation around changing inflows and electricity demand.
Approach
Chugoku Electric began exploring hydropower AI around 2019 and developed a planning system with ExaWizards. The system combined inflow predictions with mathematical optimisation. A roughly year-long trial at Sasanamigawa Dam preceded full operation at 2 dams in summer 2023.
Results
Takeaway
Forecasting inflows supported generation before incoming rain filled the reservoir. Trial results at Sasanamigawa Dam did not establish equivalent savings across the fleet.
Categories
- Industry
- Energy and utilities
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLOptimization
- Adoption
- Partner co-development
- Evidence
- Vendor case study
Sources
Source summaries
ExaWizards described jointly developed inflow forecasting and generation planning. It reported about 6 million cubic metres less annual spill in the trial and full operation at 2 dams from summer 2023.
Full details are in the original sources.
