New SAR algorithm boosts solar yield by modeling real shading patterns
Researchers use integer linear programming to dynamically adapt PV array configuration without complex switching.
Partial shading from nearby trees, buildings, or other fixed obstructions can slash solar PV plant efficiency by 20-40%. Dynamic Array Reconfiguration (DAR) can mitigate this but requires complex switching hardware and heavy computation, limiting real-world adoption. Static Array Reconfiguration (SAR) is simpler but typically assumes idealized shading regions rather than actual patterns, leaving performance on the table under time-varying conditions.
Now in a new arXiv preprint (arXiv:2606.24342), researchers Ushnik Chakrabarti and Binoy Kumar Karmakar introduce an SAR technique based on integer linear programming that uses the actual shading pattern on the PV array. By modeling the geometry of fixed obstructions and the sun’s path across seasons, the algorithm computes a one-time optimal module arrangement. Tests on square-matrix arrays in simulation and a non-square lab prototype consistently beat existing SAR methods, delivering the highest power output without any switching complexity. This approach could make high-efficiency partial-shading recovery viable for urban solar installations at near-zero hardware cost.
- Integer linear programming model captures actual shading from fixed obstructions (trees, buildings) with daily and seasonal variation.
- Outperforms existing static reconfiguration techniques in both simulation and a small-scale lab prototype.
- No complex switching hardware required – just a one-time module rearrangement based on the computed optimal layout.
Why It Matters
Brings practical, low-cost partial-shading mitigation to urban solar plants, potentially boosting energy harvest 10-30% without expensive dynamic hardware.