Performance Ratio: PR and PR Temp
Performance Ratio answers one question: of the energy the sunlight that hit your panels should have produced, how much did you actually get?
A PR of 82% means the system delivered 82% of what its sunlight, its size and its panel rating say it could deliver. The missing 18% is heat, soiling, shading, wiring, inverter efficiency and downtime.
enSights follows IEC 61724-1 for PR and the NREL weather-corrected method for PR Temp.
The timestep
Every formula below is summed interval by interval, never from one average figure for the whole period. Ξt is the interval length in hours:
Data resolution | Ξt |
|---|---|
5 minutes | 1/12 |
15 minutes | 0.25 |
30 minutes | 0.5 |
Hourly | 1 |
Intervals below 50 W/mΒ² are dropped. At dawn, dusk and under heavy cloud there is not enough light to judge a system by, so those intervals contribute to neither side of the fraction.
PR
E_AC
PR = ββββββββββββββββββββββββββ Γ 100
P_total Γ G_eff
Where the effective irradiance term is built up per interval:
G_eff = Ξ£k [ Ξt Γ (Gk / 1000) Γ (C_producing,k / P_total) ]Term | Meaning | Unit |
|---|---|---|
E_AC | Total AC energy generated in the period | kWh |
P_total | Sum of installed capacity across all devices | kWp |
Gk | Irradiance at interval k | W/mΒ² |
C_producing,k | Capacity of the devices that were producing at interval k | kWp |
Ξt | Interval length | hours |
What the capacity ratio does. The denominator only counts the sunlight that fell on equipment that was actually running. An inverter that was offline for an hour does not get charged for the sunlight it missed. This is what makes PR a measure of how well the running plant performed, rather than a measure of downtime β downtime is what Availability is for.
Curtailment. Intervals where the power plant controller was actively limiting output are excluded from the denominator entirely, so a curtailed hour neither helps nor hurts PR.
Strings or inverters. Where per-string capacity is configured, the calculation runs at string level. Otherwise it runs at inverter level. A site does the same thing for every device β it never mixes the two.
PR Temp
Panels lose power as they get hot. The same system looks worse in July than in March even when nothing is wrong. PR Temp removes that effect.
E_AC
PR Temp = βββββββββββββββββββββββββββββββ Γ 100
P_total Γ G_temp
G_temp = Ξ£k [ Ξt Γ (Gk / 1000) Γ (1 β (Tk β 25) Γ |Ξ³|) ]Term | Meaning | Unit |
|---|---|---|
Tk | Measured temperature at interval k | Β°C |
Ξ³ | Panel power temperature coefficient, from the panel model | 1/Β°C |
25 | Reference temperature | Β°C |
Ξ³ comes from the panel model configured on the system. A datasheet value of β0.35 %/Β°C is β0.0035 1/Β°C, so a panel at 45 Β°C is derated by (45 β 25) Γ 0.0035 = 7%.
Intervals with no temperature reading produce no PR Temp for the period.
PR Temp is the metric the market calls Weather-Corrected PR. It is being renamed to PR Weather Adjusted to match.
Reading the two together
Metric | Corrects for | Use it for |
|---|---|---|
PR | Irradiance, and which equipment was running | Contract baseline, IEC 61724 reporting |
PR Temp | + panel temperature | Month vs month, year vs year |
The gap between them is the temperature story: a wide gap means a hot period, not a fault.
Words to know
- PR β Performance Ratio. Actual energy Γ· energy the irradiance should have produced.
- POA β Plane of Array. Irradiance measured in the plane of the modules.
- STC β Standard Test Conditions. 1,000 W/mΒ² and 25 Β°C cell temperature, the conditions a panel is rated at.
- Ξ³ (gamma) β Power lost per Β°C above the reference temperature. Always negative on a datasheet.
- Interval (k) β One data point in time. 15 minutes for most systems.
Standards
- IEC 61724-1 β Photovoltaic system performance, Part 1: Monitoring.
- NREL/TP-5200-57991 β Weather-Corrected Performance Ratio, Kurtz et al.