PXX is a statistical notation used in the solar industry to express the probability that a solar system will produce at or above a given energy output in a given year. For example, P50 means there is a 50% probability the system will meet or exceed that production level — effectively the median expected output. P90 means there is a 90% probability the system will meet or exceed that level, making it a more conservative estimate commonly used in financial underwriting.
Understanding which PXX value Aurora uses helps you interpret simulation results accurately and align your proposals with how lenders and investors evaluate solar projects.
In this article we’ll cover:
- Key terms
- What PXX value does Aurora use for solar energy production?
- Weather datasets and P50
- P90 in battery backup durations
Key terms
- P50: The median expected annual energy production. In any given year, the system has an equal chance of producing above or below this figure. Aurora’s energy production estimates are P50 values.
- P90: A conservative estimate that the system will meet or exceed in 90% of years. P90 values are typically 5–15% lower than P50, depending on the location’s weather variability. Aurora does not natively output P90 for solar energy production, though it does use P90 in battery backup duration calculations (see Backup Duration – Calculations).
- Typical Meteorological Year (TMY): A dataset that represents average weather conditions at a location over a multi-year period. Aurora uses TMY-based weather datasets (such as NREL PSM3, TMY3, CWEC, Solcast, and IWEC2) as the basis for its simulations. Because TMY data reflects average conditions, Aurora’s output is a P50 estimate.
What PXX value does Aurora use for solar energy production?
Aurora’s solar energy production estimates are P50 values. This is because Aurora simulates system performance using Typical Meteorological Year (TMY) weather datasets, which represent average historical weather conditions at a location. A simulation run against average weather produces a median expected output — by definition, a P50.
In practice, this means:
- The annual production figure shown in Aurora represents what the system is expected to produce in a typical year.
- Actual production in any given year may be higher or lower due to real weather variability.
- If a lender or investor requests a P90 figure, you will need to apply an external P50-to-P90 adjustment factor, which is typically sourced from an independent energy assessment or your weather data provider.
Weather datasets and P50
The weather dataset you select in Aurora affects the accuracy of your P50 estimate but does not change the type of estimate — all of Aurora’s supported datasets are TMY-based and therefore produce P50 outputs.
Aurora supports the following weather datasets, selectable in Design Mode under Simulation preferences:
- NREL PSM3 and TMY3 (United States)
- Solcast and IWEC2 (International)
- CWEC (Canada)
- RMY (Australia)
For guidance on selecting or changing your default weather dataset, see Aurora Simulation Engine and How to define weather dataset default settings in new Aurora.
P90 in battery backup duration
Aurora does use P90 in one specific context: battery backup duration. When Aurora calculates how long a battery will back up a home during a grid outage, it runs the simulation across every day of the year and reports the P90 result — meaning the battery will last at least that long in 90% of scenarios. This is separate from solar energy production estimates and is documented in Backup Duration – Calculations.
Related articles