Authors: | Presented at 2025 European PVPMC workshop in Ayia Napa, Cyprus |
29 October 2025 |
Overview
As the PV industry expands into geographically challenging locations, technologies like bifacial modules, trackers, and terrain-following layouts are becoming the norm. Traditional simulation methods, specifically View Factor models, rely on simple geometric ratios. While fast, these models are unsuitable for complex shading geometries or terrain-following systems because they cannot accurately capture the intricate interplay of light and shadow in these environments.
We presented the Solargis solution to these challenges at the 2025 European PVPMC held in October 2025 at Ayia Napa, Cyprus.
Smart Ray Tracing in Solargis Evaluate
To address these limitations, Solargis has implemented Smart Ray Tracing within Solargis Evaluate. Unlike simplified models, this method simulates the physical behavior of light in a 3D environment.
Key capabilities include:
Per-Cell Resolution: The engine calculates Global Tilted Irradiance (GTI - also known as in-plane or plane-of-array irradiance) for every individual cell on both the front and rear sides of the modules.
Detailed Electrical Simulation: High-resolution optical data allows the simulator to account for specific module electrical behaviors, such as the activation of bypass diodes in half-cut modules under partial shading.
Advanced Physics: The model integrates the Perez All-Weather Sky Model for accurate diffuse light calculation and handles arbitrary shading geometries.
Quantifying the Impact
The study quantifies the error introduced by using simplified models versus detailed Ray Tracing across three scenarios, each looking at one key simplification. One year of data with 15-minute time resolution was used to run the tests across all scenarios. The results show that neglecting complex factors leads to significant deviations in yield estimation.
Scenario | Site | Studied factor | Resulting difference in specific PV output |
|---|---|---|---|
Terrain following | Japan (128 MWp) | Comparing a terrain-following layout vs. a simplified flat layout | 10.6 % relative difference |
Complex shading | Switzerland (6.3 MWp) | Accounting for external shading objects (wind turbines, buildings) and custom terrain | 3.9 % relative difference |
Tracker structure shading | Slovakia (3.3 MWp) | Accounting for the shadow cast by torque tubes on 1-axis tracker | 1.1% relative difference |
These findings demonstrate that utilizing Ray Tracing is not just a theoretical improvement but a critical requirement for accuracy, as simplified models can overestimate or underestimate yield by significant margins.
Scalability and Performance
Historically, Ray Tracing has been computationally expensive. However, this study demonstrates that cloud computing effectively removes this barrier. The scalability and performance test was run using a large 477 MWp power plant located in Egypt, using 30 years of input data with 15-minute time resolution.
The difference in the performance of the scalable cloud computing architecture of Solargis Evaluate and a high-end desktop PC is stark. While the benchmark simulation on desktop PC took 3 hours and 44 minutes to complete, Solargis Evaluate completed the same simulation in just 7 minutes - a 32x speed increase.
Conclusion and Implications
For modern PV projects, particularly those involving bifacial trackers or uneven terrain, detailed 3D optical simulation is essential. Ray Tracing provides the necessary accuracy for spatial irradiance distribution and electrical simulation. With the integration of cloud parallelization, this high-fidelity modeling is now scalable and fast enough for routine use in the design process.
For PV developers and system designers, these findings underscore the necessity of using advanced simulation tools for modern projects.
Risk Reduction: Relying on simplified terrain or shading models can lead to significant yield estimation errors (over 10% in complex terrain), directly impacting financial models and bankability.
Design Optimization: The ability to visualize per-cell GTI heatmaps allows engineers to identify hotspots and optimize stringing layouts to mitigate losses from partial shading.
Scalability: The integration of cloud computing removes the hardware bottleneck, allowing users to run complex, physics-based simulations on large-scale utility projects in minutes rather than hours or days.