How are drones used for solar panel inspection?

Timing the inspection correctly is as important as the equipment. Ideal conditions: clear sky (maximum irradiance), wind under 10 mph (thermal convection reduces hot spot visibility), panel temperature above ambient by at least 10-15C (requires at least 600 W/m2 irradiance), and no rain in previous 24 hours. In practice, this means flying 10am-2pm on clear sunny days.

Avoid partially cloudy days — cloud shadow passing over panels during flight creates false thermal readings that contaminate the data. Many inspection companies use a real-time irradiance meter and only launch when conditions meet their inspection protocol thresholds. An inspection flown in marginal conditions produces unreliable data that must be repeated.

Radiometric thermal cameras are required for quantitative solar inspection. A radiometric camera (like the FLIR Vue TZ20 or DJI Zenmuse H20T) records the actual temperature value at each pixel, not just a relative color scale. This allows post-processing software to apply consistent temperature thresholds for defect classification — for example, flagging any cell above 15C over ambient as a hot spot requiring investigation.

Without radiometric capability, two images taken at different times or camera settings are not directly comparable. Ensure any thermal drone specified for solar inspection has a fully radiometric sensor with documented NEDT (Noise Equivalent Temperature Difference) below 0.05C for reliable small-defect detection.

The ROI calculation for drone solar inspection is compelling. Manual ground inspection of a 50MW farm costs $30,000-60,000 in labor and takes 3-4 weeks. Drone inspection costs $8,000-15,000 and takes 2 days — and finds more defects since thermal is more sensitive than visual inspection.

Typical defect rates in utility-scale farms that haven't been recently inspected are 1-5% of modules with some form of degradation. A 50MW farm losing 3% of capacity (1.5 MW) at $40/MWh over 6 months loses approximately $150,000 in revenue. Finding and repairing those panels pays back the inspection cost many times over. Annual or semi-annual drone inspection is the industry best practice for farms over 10MW.

String-level vs module-level vs cell-level defect detection is a key capability discussion. From 30-50 meters altitude with a standard 640x512 thermal sensor, you can reliably detect module-level anomalies (entire panel underperforming) and string-level anomalies (entire string bypassed). Cell-level defects (small hot spots within one cell) require lower altitude — 10-20 meters — and higher-resolution thermal sensors to reliably detect.

For most O&M purposes, module-level detection is sufficient because module-level thermal anomalies indicate significant power loss requiring intervention. Finer-grained cell analysis is typically reserved for warranty disputes or root cause analysis on specific modules flagged by string-level electrical monitoring systems.

Software platforms specifically designed for solar inspection data processing significantly reduce analysis time. Raptor Maps, Solargis, and DroneDeploy Solar automatically identify thermal anomalies in processed orthomosaics, classify defects by type and severity, correlate findings with the farm's electrical layout (inverter ID, string ID, module position), and generate prioritized work order reports.

Manual review of thermal imagery from a 50MW inspection takes 40-60 hours; automated platform processing reduces this to 4-8 hours with human QC review. The time savings in data analysis often exceed the time savings in data collection when switching from manual to drone inspection. For broader thermal inspection applications, see: How are drones used for thermal inspection?