Zoltiq Battery Index
What a reference 2-hour battery earned each month on the IEX day-ahead market, planning each day on the prices it could actually see. Next to it: the best possible in hindsight, a fixed schedule that needs no software, and what trading real-time as well adds.
The Index, last month
100 MW / 200 MWh battery, one cycle a day, after exchange fees and wear. IEX prices through the latest day.
The Index data could not be loaded. Please try again later.
Earnings per MW, month by month
₹ lakh per MW of battery power earned in each calendar month. Hover or tap for values.
Every battery, every strategy
The same battery run through every schedule we test: two sizes, both markets, one or two cycles a day, with or without a rule that only allows charging in the solar hours. All without look-ahead, except the hindsight rows, which are ceilings.
Month by month
₹ lakh per MW earned in each month by four of the schedules above.
Year by year
The strategies
- Perfect hindsight
- The best schedule possible with every price of the day known in advance. A ceiling no one can reach.
- 7-day average plan
- Each day is planned on the average of the last 7 days' prices that were public at bid time, by an exact optimiser. This is the Index.
- Yesterday's prices
- The same optimiser, planning on the most recent day's prices.
- Same day last week
- The same optimiser, planning on the same weekday one week earlier.
- Monthly windows
- One charge window and one sell window picked at the start of each month from the last 90 days, then repeated every day.
- Fixed 11:00 / 18:00
- Charges from 11:00 and sells from 18:00 every day, for the battery's duration. No software at all.
- Day-ahead plan, corrected in real-time
- Both markets. At noon the day before, commits the 7-day average plan in day-ahead. At 22:00, when the next day's day-ahead prices are public, re-plans on a blend of those prices and last week's real-time prices, and buys or sells the difference in real-time.
- Best market per block, chosen ahead
- Both markets. At noon the day before, buys each block in whichever market last week's prices say is cheaper, and sells in the dearer.
- Best of both markets
- Hindsight, buying in whichever of day-ahead and real-time is cheaper in each block and selling in the dearer. An upper bound.
- Lease at the tender tariff
- The alternative to trading: rent the battery to a discom at the latest tender tariff for its duration. Fixed, and before the buyer's charging costs.
Against tender tariffs
Recent standalone battery tenders, in the same unit.
How it is calculated
Every value is reproducible from IEX's public 15-minute data with the open backtester.
Three steps
- Plan. At noon on the day before delivery, when day-ahead bids close, the battery sees only the prices already published. It averages the last 7 days block by block and an exact optimiser picks when to charge and discharge on that profile.
- Settle. The plan is settled at the day's actual clearing prices, block by block, with exchange fees on both sides and a wear charge on every MWh discharged. Blocks where nothing traded do not fill.
- Add up. Each day's profit is added into its calendar month and divided by the battery's power: ₹ per MW per month, the unit Indian battery tenders use.
Assumptions
| Battery | 100 MW / 200 MWh (2-hour); 100 MW / 400 MWh (4-hour) |
|---|---|
| Round-trip efficiency | 87%, split evenly between charging and discharging |
| State of charge | 10–90%; each day starts and ends at 10% |
| Cycles | 1 a day (Index); 2 as a variant |
| Market | IEX day-ahead (Index); real-time as a variant |
| Exchange fee | ₹20/MWh on each side |
| Wear | ₹300 per MWh discharged |
| Market share | At most 10% of a block's cleared volume |
| Not included | Transmission charges, O&M, deviation charges, ancillary income, capacity fade |
Data
Every monthly value since May 2022. Cite as: Zoltiq Battery Index, zoltiq.in/battery-index, accessed [date].
Backtested results are not a guarantee of future returns. The battery is assumed not to move prices, which holds for a 100 MW battery but not for a large fleet.