Argus PV simulation chain

In this document

This document provides a high-level introduction to the Argus PV simulation engine used in Solargis Evaluate. It describes the purpose and structure of the simulation, summarizes each stage of the simulation chain, and explains where to find detailed methodology documentation for each stage.

Overview

The Argus PV simulator is Solargis's most advanced energy yield simulation engine, integrated into Solargis Evaluate. It estimates the electricity output of a PV power plant based on the site's solar radiation data, meteorological conditions, environmental factors, and the full technical configuration of the energy system.

Argus sets a new standard for PV simulation accuracy by combining technologies and models that are not available together in any other commercially available simulator. Industry-standard tools such as PVsyst and SAM use view-factor models, which assume isotropic scattering of reflected rays. Argus goes further: it employs unbiased Monte Carlo backward raytracing in combination with Perez's all-weather sky model to simulate light at the individual PV cell level — accounting for direct, diffuse, and reflected radiation with a level of precision that view-factor approaches cannot achieve. This enables accurate shading simulation for complex layouts, including partial shading of individual cells and gaps between cells.

Beyond raytracing, Argus incorporates several capabilities to stand out in the field of solar energy simulation:

Proprietary physics-based soiling model — dynamically estimates soiling losses from atmospheric pollution data, rainfall, wind, and module tilt for any location worldwide, rather than relying on user-specified monthly percentages.
Snow loss simulation — integrates the Solargis snow loss model to quantify energy losses from snow coverage, a factor not natively modeled in most simulation tools.
Transient thermal correction — accounts for the thermal inertia of PV modules using a weighted moving-average temperature model, improving accuracy under rapidly changing irradiance conditions.
Spectral correction — applies the Lee & Panchula model based on air mass and precipitable water content, capturing the effect of atmospheric composition on module spectral response.
Cell-level IV curve simulation — calculates current-voltage (IV) curves for every PV cell in the power plant using De Soto's single diode model, eliminating the need for simplified partial-shading assumptions required by other tools.
Full bifacial simulation — performs separate front- and rear-side raytracing for bifacial energy systems, applying the bifaciality factor from the modules loaded from PV Components Catalog (PVCC).
IEC 61724-3 compliance — the implementation of technical unavailability losses follows IEC Technical Specification 61724-3, ensuring results meet international standards for energy evaluation of photovoltaic systems.

Flowchart illustrating inputs and computations for evaluating PV simulation processes.

Argus PV simulator - the simulation stages

The simulation chain is organized into four sequential stages, each building on the results of the previous one.

Stage 1: Simulation inputs

Simulation inputs are the foundation of the Argus simulation chain. Before any calculation begins, the simulator collects all data that defines the site conditions and the energy system configuration. The accuracy of these inputs determines the reliability of every result that follows.

Sun geometry — The solar position (azimuth and elevation) is calculated using the PSA model from geographical coordinates and date-time information. Accurate sun position is required for all shading, irradiance, and optical calculations.
GHI and DNI — Global horizontal irradiation and Direct normal irradiation are provided as Solargis Time Series or Typical Meteorological Year (TMY) data at 15-minute or 1-minute resolution. These are the primary solar radiation inputs that drive the entire simulation.
Terrain and horizon — Terrain is sourced at 30 m or 90 m spatial resolution and treated as a shading object; the Solargis-derived horizon is provided at 7.5-degree horizontal resolution. Both determine how far-field and near-field shading is applied to each segment.
Albedo — Monthly ground albedo and object albedo values are configured per project and affect the reflected radiation component reaching module surfaces, including rear-side irradiance in bifacial systems.
Meteorological parameters — Parameters including air temperature, wind speed, atmospheric pressure, precipitable water, precipitation, dust, and snow are supplied as time series. These drive thermal, spectral, soiling, and snow loss models.
Energy system configuration — The physical layout (mounting system, spacing, shading objects) and electrical layout (PV modules, strings, inverters, transformers, grid connection) are defined in the Energy system designer. All system elements participate in shading and electrical calculations.
PV module and inverter specifications — Verified component parameters are sourced from the PV Components Catalog (PVCC), ensuring consistent, high-quality inputs for the electrical simulation.

For full details on simulation inputs, see Argus simulation inputs.

Stage 2: Optical simulation

The optical simulation calculates how much solar radiation reaches each PV module surface under real-world conditions, applying all optical losses in sequence. Its output — the spectrally corrected Global tilted irradiance (GTI) per cell — is the direct input to the electrical simulation.

Sky irradiance model — Perez's all-weather sky model distributes diffuse radiance across the sky dome using GHI and DNI inputs, and the theoretical no-loss GTI is calculated from these distributions. This model is widely recognized as the most accurate for anisotropic sky conditions and is the basis for all subsequent irradiance calculations.
Far horizon shading — Horizon shading is applied using a View Factor model, with the far horizon pre-loaded from Solargis data for each segment's reference point. This removes irradiance blocked by distant terrain features before near-field calculations begin.
Near shading — 3D backward raytracing — All system objects and the surrounding terrain are placed in a 3D calculation scene and simulated using unbiased Monte Carlo path tracing, separately for front and rear module surfaces. This is the most computationally rigorous shading method available, enabling cell-level accuracy and correct modeling of partial shading without simplifying assumptions.
Soiling losses — The Solargis proprietary soiling model estimates soiling losses based on atmospheric particle concentration, wind, and rainfall, applied as monthly or yearly attenuation to the GTI for both module faces. Accurate soiling estimation is essential for sites in arid or polluted environments, where losses can exceed 20%.
Angular reflection losses — The Martin and Ruiz model accounts for optical losses caused by the angle of incidence on the module surface, using an angular loss coefficient estimated from the module's optical properties. These losses increase at low sun angles and affect the total radiation reaching the PV cell.
Spectral correction — The Lee & Panchula model corrects for the spectral mismatch between incident sunlight and the module's spectral responsivity, using air mass and precipitable water content as inputs. This correction captures the effect of atmospheric composition on energy conversion efficiency, which varies significantly by climate and season.

For full details on the optical simulation, see Argus optical simulation.

Stage 3: Electrical simulation

The electrical simulation converts the spectrally corrected GTI from the optical stage into AC electrical power at the grid connection point, accounting for all electrical losses along the path from PV cell to grid. It uses component-level models throughout, eliminating the need for simplified loss assumptions.

Conversion of irradiation to DC electricity — De Soto's single diode model generates IV curves for every PV cell in the power plant using five electrical parameters from the PVCC and the cell temperature from the transient thermal correction model. This cell-level approach allows the simulator to capture partial-shading effects on string performance without generalized mismatch correction factors.
Inverter clipping losses — When the DC array output exceeds the inverter's rated AC capacity, the excess power is clipped and lost. Clipping losses are calculated from the string IV curves at the inverter input, reflecting real operating conditions at every time step.
Grid power limitation losses — If the grid operator imposes a power injection limit below the plant's rated output, the inverters curtail their output accordingly. This loss is applied directly from the grid connection settings defined in the Energy system designer.
DC losses — Resistive losses in DC cables and combiner boxes are applied as a percentage of DC power at Standard Test Conditions (STC). The default value in Solargis Evaluate is 2%, adjustable in the Cabling section of the Energy system designer.
Inverter DC/AC conversion — The Sandia inverter model converts DC input to AC output by determining inverter efficiency as a function of operational and environmental conditions, and calculates active and reactive power components based on the user-specified power factor. Losses including night standby consumption are taken from the PVCC.
Auxiliary losses — Energy consumed by monitoring systems, tracking drives, lighting, heating and cooling, and other auxiliary equipment is deducted as constant (day and night) and proportional losses. These losses are separate from inverter and transformer losses.
Transformer losses — Solargis's proprietary transformer model accounts separately for no-load (iron) losses and load (copper) losses for each transformer stage, covering both inverter transformers and power transformers. Losses are computed for each transformer individually based on its power rating and settings.
AC losses — Resistive losses in AC cabling are applied at low-voltage (LV), medium-voltage (MV), and high-voltage (HV) stages depending on the number of transformer stages in the power plant. Default values are 1%, 0.5%, and 0.05% respectively, all adjustable in the Energy system designer.

For full details on the electrical simulation, see Argus electrical simulation.

Stage 4: Post-processing

Post-processing is the final stage of the Argus simulation chain, applied after the electrical simulation has produced AC power output at the grid. It accounts for system-level factors that affect long-term energy delivery and cannot be modeled within the electrical simulation itself.

System unavailability losses — Electricity losses from unplanned equipment failures, scheduled maintenance, and external events such as grid curtailment are quantified as internal and external unavailability, respectively. These losses are implemented in accordance with IEC Technical Specification 61724-3, which defines energy evaluation methods for photovoltaic systems.
Snow losses — The Solargis snow loss model estimates energy losses caused by snow coverage of PV modules, using meteorological time series data. Snow losses are location-specific and can be significant in high-latitude or high-elevation sites.
Long-term degradation — The performance reduction of PV modules and other components over time is modeled over a 25-year horizon, applying separate degradation rates for the first year and subsequent years. Degradation is applied to the long-term average PVOUT and Performance Ratio (PR) derived from the simulated historical data.

For full details on post-processing, see Argus post-processing.