SolarFarmer and PVsyst Differences
Modelling differences
The models used in SolarFarmer to estimate all effects in a PV system represent DNV best practices. When comparing different individual effects in SolarFarmer and PVsyst, they can differ because both packages use either different models or have differences in the implementation.
The table below lists major differences in SolarFarmer and PVsyst as known to date.
Irradiance
| Model | SolarFarmer | PVsyst |
|---|---|---|
| Solar position | NREL Solar Position Algorithm (SPA) [10] | Simplified expressions of the U.S. Naval Observatory Solar Coordinates Model as proposed by the Ineichen model [28] |
| Ground reflected irradiance | Considers inter-row albedo contribution in both, monofacial and bifacial modules | Does not consider inter-row albedo contribution to irradiance for monofacial modules |
| Horizon | Considered for direct irradiance only. Always calculated. | Considered for direct, sky diffuse and reflected irradiance components. Not calculated when all horizon is 2 degrees or smaller elevation. |
| Near shading | 2D model differentiates between first, middle and last rows | Unlimited sheds model treats all rows the same |
| IAM diffuse | Diffuse components require integrating the IAM curve but implementation differs in SolarFarmer and PVsyst | Diffuse components require integrating the IAM curve but implementation differs in SolarFarmer and PVsyst |
| IAM back side | Uses IAM model defined for module type | Uses Fresnel model for glass without anti-reflective coating |
Tracking
| Model | SolarFarmer | PVsyst |
|---|---|---|
| Tracker angles | True-tracking, NREL standard backtracking, slope-aware backtracking or user-defined custom tracking angles. | Astronomical with or without backtracking |
Conversion
| Model | SolarFarmer | PVsyst |
|---|---|---|
| Electrical mismatch due to shading | Full calculation at submodule level | Fraction loss for electric effect (user input, e.g. 70 %) |
Inverter
| Model | SolarFarmer | PVsyst |
|---|---|---|
| Maximum power (clipping) | Determined based on AC rated power after applying all DC losses and inverter DC/AC efficiency | Determined on DC side, converting AC rated power to DC by applying DC/AC efficiency. As a result, the maximum DC current limiting loss is very often "preceded" (i.e. masked) by the maximum power loss |
Effects reporting
Loss/gain factors are a comparison of the energy before and after a specific effect is applied. This means that reported effects depend on the order in the calculation chain. The calculation order in SolarFarmer differs from PVsyst, which makes individual effects look different. However, this is only caused by a different reference value being used for the effects' percentages. The calculation order is shown below.
It should be noted that the calculation order is different in the SolarFarmer local and API calculations.
| SolarFarmer local | SolarFarmer API | PVsyst |
|---|---|---|
| Irradiance | ||
| Tilt | Tilt | Tilt |
| Horizon | Horizon | Horizon |
| Near shading (front side) | Soiling | Near shading (front side) |
| Soiling | IAM | IAM |
| IAM | Spectral | |
| Spectral | Backside including IAM | Soiling |
| Backside including near shading | Backside structural shading | |
| Backside structural shading | Backside transmission | |
| Backside transmission | Near shading (front and back) | Backside including all effects |
| Conversion | ||
| Temperature | Irradiance | Irradiance |
| Irradiance | Temperature | Temperature |
| Bifaciality factor | Module power binning | Spectral |
| Back irradiance mismatch | LID | Electrical mismatch |
| Module power binning | Module Quality | Module Quality |
| LID | Module mismatch | LID |
| Module Quality | Bifaciality factor | Module mismatch |
| Module mismatch | Back irradiance mismatch | |
| Electrical mismatch | Electrical mismatch | |
| DC collection | DC collection | DC collection |
| Inverter | ||
| Min DC Voltage | Min DC Voltage | Efficiency |
| Max DC current | Max DC current | Max DC power |
| Max DC voltage | Max DC voltage | Max DC current |
| Min DC Power | Min DC Power | Max DC voltage |
| Efficiency | Efficiency | Min DC power |
| Max AC power | Max AC power | Min DC voltage |
| Over power shutdown | Over power shutdown | Tare |
| Tare | Tare | |
| Collection | ||
| AC collection | AC collection | Auxiliaries consumption |
| Transformer | Transformer | AC collection |
| Grid connection limit | Grid connection limit | Transformer |
| Grid availability | Grid availability | Grid availability |
The local and API codebase differ so running the same site locally or in the Cloud may not give the same results. Some of the differences are:
- Backside irradiance
- Shading models differ
- No IAM applied on back side in local calculation
- 2D model assumes horizontal terrain while API supports tilted plane
- Local does not support treating circumsolar as direct irradiance, it always treats it as diffuse
Getting close to PVsyst results
Given the differences in the models and their implementation, it is expected that SolarFarmer and PVsyst results are not identical. It is possible the get results close to the PVsyst unlimited sheds simulations when using the SolarFarmer 2D API.
Some additional adjustments of the JSON file produced by the SolarFarmer Desktop may be needed when calling the API.
EnergyCalculationOptions
useMostShadedCellForDiffuse=True
modelBackRowSeparately=False
Layout
Fixed tilt:
numberOfStringsInFrontRow=0
Trackers:
numberOfStringsInRightRow=0
numberOfStringsInLeftRow=0