Greenbyte Documentation

Potential Power Learned
Description

Potential Power Learned is a purely data-based method to estimate the ideal potential production for an inverter. Based on historical data, a model is created to replicate the best performance of an inverter for different operating points. The main benefit of Potential Power Learned is that it does not require any metadata to work but is still able to estimate potential power accurately.

Potential Power Learned requires configuration. If you are interested in using it, please contact Greenbyte Support.

Technology

Solar: Inverter

Equation
potential_power_model_simulated_eq.jpg

where:

Pp

is Potential Power Learned

Ic

is Solar Irradiance Reference

Cut-inIA

(optional) is the Availability irradiance cut-in (W/m2) (Inverter metadata) applied to the irradiance

Ta

is the average ambient temperature from reference pyranometers

a, b, c, d

are coefficients calculated based on historical data

Inputs
potential_power_model_simulated_in.jpg
Example

In this example, the Potential Power Learned is compared to Potential Power Ambient Temperature and you can see that Potential Power Learned is much closer to the actual power being produced by the inverter.

potential_power_model_simulated_ex.jpg