Linear Programming optimisation run on historical prices, co-optimising energy arbitrage and FCAS on one shared battery. Run two ways: with perfect foresight (the theoretical ceiling) and on AEMO's public price forecast (the forecast-informed benchmark). Reference points for actual performance.
The LP backcast is the linear-programming optimisation NEMPulse runs over historical AEMO price data to establish what a given battery could have earned under an idealised dispatch strategy, co-optimising energy arbitrage and all ten FCAS markets together on one shared unit so a megawatt committed to FCAS is correctly unavailable for arbitrage in that interval.
It is run in two configurations. The perfect-foresight run assumes every future price is known in advance and produces a theoretical ceiling that no real operator can reach. The forecast-informed run instead solves against AEMO's own public PREDISPATCH price forecast — issued before each trading day, with no hindsight — then settles at the prices that actually occurred, isolating the value of price information alone while holding execution flexibility equal to the perfect-foresight case. Both are subject to the same operational constraints as the real battery: state-of-charge limits, a per-day cycle cap, and FCAS response obligations.
Related terms: Energy CaptureForecast Benchmark
See also: Actual vs optimal performance
Explore: Full glossary
Loading live data…