For most of the last decade, the problem with Indian solar was getting enough of it built. That problem is now being solved faster than the grid around it.
The result is curtailment: grid operators instructing solar and wind plants to produce less than they could, even on a clear day. The energy is not stored or sold somewhere else. It is simply never generated.
How big the problem is
The think tank Ember estimates that India curtailed about 2.1 TWh of renewable generation in FY2025-26 (April 2025 to March 2026). That is roughly 1.3 per cent of renewable output across the year — but it is concentrated in the middle of the day, and by April 2026 curtailment during peak solar hours had reached about 4 per cent. An earlier Ember estimate, using a different period and method, put solar curtailment for May to December 2025 at around 2.3 TWh, so treat any single figure as indicative rather than exact.
The trend matters more than the level. India added about 24 GW of solar between October 2025 and April 2026 alone, bringing the total to roughly 154 GW, and every gigawatt added at midday makes the midday squeeze tighter.
Cause one: coal cannot turn down any further
At noon on a sunny day, solar output can exceed what the grid needs. Other generators must back off to make room. But India’s coal fleet can only reduce output so far — to a minimum technical load of about 55 per cent of rated capacity. Below that, units become unstable or have to shut down entirely, and they cannot then restart in time for the evening peak.
Ember’s analysis shows how hard the system is already being pushed. On 6 March 2026, solar and wind supplied 41 per cent of generation at midday, forcing coal to drop 49 GW in six hours and then climb 51 GW by the evening. By April 2026, coal had breached its minimum technical load floor in more than half of all midday dispatch intervals, and curtailing renewables supplied 37 per cent of the downward adjustment that month.
When coal hits its floor and supply still exceeds demand, the only thing left to cut is renewable output.
Cause two: the plant is ready, the line is not
The second cause is local. In Rajasthan, where much of India’s utility solar is being built, nearly 4 GW of wind and solar capacity was reported curtailed from March 2025, as transmission projects ran late and lower-than-expected demand during a wet monsoon compounded the problem. Developers at the time expected it to continue for months.
This is not unique to one state. New plants in Rajasthan, Gujarat, Maharashtra and Tamil Nadu have been commissioned on only temporary network access, and industry reporting describes transmission build times stretching from about two years to as much as three. Down To Earth reported that at least 30 solar and wind projects faced curtailment between March and August 2025, with some instructed to cut up to 48 per cent of daily generation during constrained periods. We explain how access is granted in GNA connectivity and storage.
Where storage helps — and where it does not
For midday oversupply, storage is the most direct fix. A battery charges from the surplus that would otherwise be thrown away and discharges into the evening, when India’s real shortage sits — the gap we cover in India’s evening peak storage gap. Ember estimates that around 10 GWh of storage would have avoided essentially all of the curtailment it measured in FY2025-26. That is a modest amount compared with the pipeline already tendered.
For missing transmission, storage helps but does not replace the line. A battery placed behind a congested connection can soak up output during the hours the line is full and export it when there is room. That raises the energy a plant can actually deliver and is part of why planners are now treating storage as a transmission asset. But if a line is simply not built, a battery can only shift energy within the day — it cannot deliver more energy than the connection can carry over 24 hours.
Policy is already moving
In February 2025, the Ministry of Power issued an advisory asking renewable energy implementing agencies and state utilities to include co-located storage of at least 2 hours, equal to 10 per cent of installed solar capacity, in future solar tenders. It also suggested distribution companies consider similar requirements for rooftop solar. The government expected around 14 GW / 28 GWh of storage from this route by 2030.
That sits alongside the firm-and-dispatchable tender formats we explain in FDRE tenders, which require developers to deliver a shaped, reliable profile rather than whatever the sun produces.
Curtailment rules, access terms and tender storage requirements change by notification and regulatory order. Treat the figures here as a September 2026 snapshot and verify the current terms before making investment decisions.
What this means for you
- If you own or are building a solar plant, look hard at your connectivity type and your curtailment history. If you are on temporary access or in a congested region, a co-located battery is a hedge against lost generation, not just an add-on. Our solar + BESS configurations are designed for exactly this.
- If you are bidding for new solar tenders, assume storage will be a requirement rather than an option, and price it in from the start.
- If you are a C&I buyer with rooftop or captive solar, the same logic applies at a smaller scale: storing midday output you cannot use is usually worth more than exporting it at a low rate.
If you want to work out how much curtailed or surplus energy a battery could recover at your site, talk to our team.