PRESS RELEASE FOR IMMEDIATE RELEASE
Solar and batteries can run like a thermal plant, at less than the price of new coal power, UC Berkeley study finds
An analysis of India’s landmark 1,000 MW “thermal mimic” auction finds that about 3 GW of solar and 12 GWh of batteries per 1,000 MW contracted can serve the tender’s round-the-clock profile at ₹5.15 per kWh, inside the ₹5.25 discovered tariff and below recent thermal power purchase agreements of ₹5.4 to 6.3.
On August 7, the Solar Energy Corporation of India (SECI) concluded the reverse auction for its 1,000 MW FDRE-RTC-V tender, the first tender in the world to require renewable energy to deliver a thermal power plant’s shaped operating profile, block by block. Earlier round-the-clock procurements, in India and abroad, bought flat baseload supply or annual energy volumes; none dispatched renewables to the shape of the grid’s actual need. The contract requires, in every 15-minute block of the year, at least 90 percent of contracted capacity during six night-and-morning peak hours chosen daily by the buyer, 50 to 60 percent during solar hours, and at least 70 percent in all remaining hours, with shortfalls penalized at 1.5 times the tariff. Seven developers won at ₹5.25 to 5.26 per kWh, a price India’s Ministry of New and Renewable Energy noted is cheaper than recent thermal PPAs at ₹5.4 to 6.3 per unit.
The profile the tender buys is, in effect, India’s net load: the demand that remains after cheap midday solar floods the grid. It is the shape India’s coal fleet already runs today, backing down toward minimum load through the daytime and ramping up as the sun sets. SECI wrote that operating reality into the contract. The tender takes only 50 to 60 percent of contracted capacity during solar hours, because utilities argued they should not pay round-the-clock prices for daytime energy they can already buy from plain solar plants at around ₹2.5 per kWh, while demanding 90 percent through the evening and night peaks.
That inversion makes the economics harder for developers, not easier. The plant’s cheapest and most abundant energy, midday solar, is exactly what the contract caps: anything above the 60 percent daytime limit earns only half the tariff. The price must instead be carried by energy shifted through batteries into the night, which is why the ₹5.25 result is, above all, a signal of how cheap storage has become.
A new University of California, Berkeley study, based on hourly simulation of the final tender rules across ten states and ten weather years, finds the winning price is deliverable with today’s technology. The least-cost design is about 3.0 GW-AC of solar (4.2 GW-DC) paired with 12 GWh of usable battery storage for every 1,000 MW contracted, at a capital cost of roughly ₹22,650 crore per GW. Sited in Rajasthan, that plant serves the full profile at an all-in cost, penalties included, of ₹5.15 per kWh.
The study finds the reliability of these plants is thermal-equivalent in the fullest sense. In a typical weather year, the design meets its supply floors outright in 94 percent of all hours, and the tender’s allowance for purchasing up to 5 percent of required energy from the green market covers nearly all of the remainder, leaving penalties of about half a percent of revenue. Even in the worst weather year of the decade simulated, penalties stay near 3 percent of revenue. And for developers who want no penalty exposure at all, an additional half gigawatt of solar eliminates penalties entirely in every weather year at an all-in cost of ₹5.28 per kWh, still within the discovered tariff.
The comparison with coal is direct. Scoring India’s 213 operating coal stations against the same profile, using their actual outage records from June 2024 to May 2025, the study finds 48 percent of stations would have paid penalties under the tender’s rules, and 18 percent would have lost more than a tenth of their revenue. The solar-plus-storage plants designed in the study are more reliable, under this tender’s own standard, than most of the thermal fleet they are asked to imitate.
“SECI asked renewables to behave like a thermal plant, and the market answered at a price below new coal. Our simulations show that this is not aggressive bidding but sound engineering: sized correctly, solar and storage now hold a thermal profile through monsoons and winter fog, with penalty exposure smaller than the coal fleet’s own outage record,” said Umed Paliwal of the University of California, Berkeley, the study’s author.
The result implies installed battery system prices near $93 per kWh, well below India’s standalone storage benchmarks but consistent with DC-coupled plants commissioning in 2028. Because every $10 per kWh decline in battery prices lowers the delivered cost by about ₹0.28 per kWh, the study projects that future round-the-clock auctions could clear near ₹4.3 to 4.6 per kWh as battery prices continue to fall, decisively below the cost of new coal power.
Methodology. The study encodes the final tender rules (RfS and Amendments 1–3) and simulates hourly dispatch of 64 solar-battery configurations in each of ten states over weather years 2015–2024, using NREL solar resource data, a 3 percent equipment failure rate on all generation, and conservative Indian cost benchmarks (solar at ₹3.75 crore per MW-AC, batteries at ₹8,550 per kWh, WACC of 10 percent). All quoted costs include penalties as simulated. Coal stations are scored from Central Electricity Authority daily outage data at station level, with the same market-energy allowance.
Contact. Umed Paliwal, University of California, Berkeley, umed@berkeley.edu
MEDIA KIT
Note: the two working papers below are separate studies from this tender analysis — they evaluate a different plant (5 GW solar + 16 GWh storage) serving a flat coal-equivalent profile, not the tender’s shaped contract.
RTC-V Analysis Deck (PDF)
Full slide deck on SECI's FDRE-RTC-V auction: the delivery profile, least-cost plant design, reliability and penalty exposure, and the coal-fleet comparison.
1.4 MB PDF
RTC-V Analysis Deck (PPTX)
Editable PowerPoint version of the RTC-V analysis deck, for reuse of individual slides and charts.
1.2 MB PPTX
RTC-V Chart Pack (ZIP, 21 figures)
Every RTC-V figure as a print-resolution PNG, for reuse in reporting and presentations.
2.1 MB ZIP
Paper 1 — Plant-Level Evidence (PDF)
From Coal-Equivalent Plants to a Coal-Equivalent Fleet: Plant-Level Evidence for Firm Solar in India. Citable working paper.
3.9 MB PDF
Paper 2 — Continental-Scale Fleet (PDF)
From Coal-Equivalent Plants to a Coal-Equivalent Fleet: Geographic Diversification at Continental Scale. Citable working paper.
6.3 MB PDF
Media enquiries
Umed Paliwal, University of California, Berkeley. Interviews, data requests, and figure reuse.
umed@berkeley.edu email