Direction 01
A rooftop-solar advisor for Chennai apartments: photos plus bill history in, honest sizing and payback ranges out
Sustainable Development Goal 07 / 17
Affordable and Clean Energy
Ensure access to affordable, reliable, sustainable and modern energy for all. For builders: tools that help households, institutions and small industries consume smarter and adopt clean energy with confidence.
Why this goal matters
The energy transition is decided in millions of small decisions: whether a family trusts rooftop solar, whether a small factory can see which machine wastes power, whether a village microgrid stays maintained. Software that turns invisible energy flows into understandable decisions accelerates all of it.
The Indian context
India is simultaneously one of the world’s largest renewable-energy builders and a country where electricity reliability still varies widely. Rooftop-solar policy support exists but adoption is slowed by trust, sizing confusion and paperwork. Small and medium industries — a huge share of national consumption — rarely have energy-analytics access that large factories take for granted.
Tamil Nadu, specifically
Tamil Nadu is an Indian wind-power pioneer — the Muppandal belt near the southern tip is among Asia’s largest wind concentrations — and a major solar state. Coimbatore’s pump, motor and textile industries make industrial energy efficiency a natural local problem; Chennai’s apartment towers make rooftop-solar decision tools immediately useful.
People and communities affected
- Households facing rising tariffs with no usage visibility
- Small industries with energy-hungry legacy equipment
- Farmers running irrigation pumps on subsidised power
- Apartment associations weighing solar investments
- Off-grid and unreliable-grid rural communities
Key challenge areas
- Consumption visibility and waste detection
- Rooftop-solar decision support and sizing
- SME energy-efficiency analytics
- Renewable-asset maintenance prediction
- Demand planning for local grids
The AI Lens
Where AI belongs here — and where it does not.
Where AI can meaningfully help
- Disaggregating a building’s load to show which equipment consumes what
- Solar-yield estimation from roof geometry, shading and local weather history
- Predictive maintenance signals for wind and solar assets from sensor patterns
- Tariff-aware scheduling suggestions for flexible industrial loads
Where AI may be inappropriate
- Guaranteeing financial returns on energy investments — models inform, they do not promise
- Automated control of safety-critical electrical equipment without certified engineering
- Punitive consumption profiling of low-income households
Build Directions
Example problem directions
Starting points, not prescriptions — the strongest submissions narrow one of these into a specific, evidenced local problem.
Direction 02
An SME energy copilot that finds the three cheapest fixes from a month of smart-meter data
Direction 03
A pump-efficiency assistant for farm clusters comparing similar pumps to spot degradation
Direction 04
A hostel or campus energy dashboard that turns consumption into weekly actions
Potential users
- Households and apartment associations
- SME owners and energy auditors
- Renewable-asset operators
- Campus facility teams
Possible datasets
- Your own metered pilot data (smart plugs, meters) with consent
- NREL/NASA solar irradiance datasets (public)
- CEA and TANGEDCO published generation and tariff information
- Open building-load research datasets for pre-training
Data-access limitations
- Indian appliance-level load signatures differ from Western research datasets
- Utility meter data access requires the customer’s own download or consent
- Weather-to-yield models need local calibration to be honest
Privacy, bias and safety risks
- Over-promised savings destroy trust in the whole category
- Electrical-safety advice must defer to qualified electricians
- Optimisation that ignores comfort or production reality gets uninstalled
Responsible-AI considerations
- Present estimates as ranges with assumptions visible
- Keep humans in the loop for any physical control action
- Make the baseline measurable before claiming improvement
Suggested impact metrics
- Measured kWh reduction against a defined baseline in the pilot
- Estimate accuracy: predicted versus actual solar yield or savings
- Payback-model transparency: every assumption inspectable
Related SDGs
Related challenge areas
Participant resources
Build for Affordable and Clean Energy.
Anchor your team in this goal, define one real problem, and start the three-lock journey.
Register Your Team