Perfect foresight · test window 1 Jul 2021 – 11 Aug 2021

Bigger, longer, leakier

The brief fixes the battery at 300 MW and 580 MWh with 90% efficiency each way. These sweeps re-solve the six-week optimisation for other designs, so you can see which knob matters.

Double the energy

+38%

300 MW with 1,160 MWh earns $8.19M vs $5.93M.

Double the power

+35%

600 MW with the same 580 MWh earns $7.99M: it can fill and empty faster, chasing shorter spikes.

A lossless battery

+26%

100% round trip instead of 81% adds $1.55M and 54 more cycles.

Power × energy

Six-week revenue by battery size

Perfect-foresight test-window revenue for each power and capacity
MWPower ↓ / Energy →	150 MWh	290 MWh	580 MWh	870 MWh	1,160 MWh	2,320 MWh
75 MW	$1.51M2.0 h · 122 cyc	$2.05M3.9 h · 85 cyc	$2.46M7.7 h · 50 cyc	$2.62M11.6 h · 35 cyc	$2.72M15.5 h · 26 cyc	$3.05M30.9 h · 13 cyc
150 MW	$2.04M1.0 h · 187 cyc	$2.96M1.9 h · 124 cyc	$4.09M3.9 h · 85 cyc	$4.65M5.8 h · 63 cyc	$4.93M7.7 h · 50 cyc	$5.45M15.5 h · 26 cyc
300 MW	$2.67M0.5 h · 295 cyc	$3.99M1.0 h · 191 cyc	$5.93M1.9 h · 124 cyc	$7.26M2.9 h · 100 cyc	$8.19M3.9 h · 85 cyc	$9.86M7.7 h · 50 cyc
450 MW	$2.73M0.3 h · 316 cyc	$4.63M0.6 h · 247 cyc	$7.08M1.3 h · 158 cyc	$8.89M1.9 h · 124 cyc	$10.3M2.6 h · 106 cyc	$13.5M5.2 h · 69 cyc
600 MW	$2.73M0.3 h · 316 cyc	$5.20M0.5 h · 301 cyc	$7.99M1.0 h · 191 cyc	$10.1M1.4 h · 147 cyc	$11.9M1.9 h · 124 cyc	$16.4M3.9 h · 85 cyc

Single six-week totals. For intervals over days, paired changes and elasticities, see sensitivity with uncertainty. Rows are power (MW), columns energy (MWh). Outlined: the brief's 300 MW / 580 MWh battery ($5,926,534). Hours = energy ÷ power. Revenue rises with both, but with diminishing returns: the deepest troughs and highest peaks are used first.

Losses

Round-trip efficiency

round-trip efficiency (split evenly between charge and discharge)

Every lost percentage point shrinks the set of price gaps worth trading, so revenue and cycling fall together.

Wear

Degradation cost per MWh delivered

$ per MWh delivered to the grid (the brief assumes $0)

A $10/MWh wear charge trims revenue by 10% but cuts cycling by 18%: the optimiser drops the marginal round trips first. Try it on a single day in the Dispatch Lab.