Drone thermography detects overheating electrical equipment, high-resistance connections, and process vessel hot zones — all leading indicators of industrial fire risk. Early thermal detection allows intervention before a hotspot escalates to a fire event.
Why This Guide Exists
We have delivered drone thermography reports to plant managers, O&M contractors, EPC engineers, and asset managers across India. The most common question after delivery is not what did you find — it is what does this actually mean and what do I do next. This guide walks through every section of the Elion report, explains what each element means, and gives clear guidance on the appropriate response to each severity class.
Section 1: Inspection Conditions — What to Check First
The first section of every Elion report documents the conditions under which the inspection was conducted. This matters because the reliability of thermal anomaly detection depends directly on irradiance, wind, and module operating temperature.
| Parameter | Recorded by Elion | Why It Matters |
| Date and time | Start and end for each inspection zone | Allows correlation with plant SCADA generation data on that day |
| Irradiance | Measured in W/m2 at ground level every 30 minutes | Must be above 600 W/m2 for IEC 62446-3 compliance — lower irradiance reduces fault detectability |
| Wind speed | Measured at inspection height | Above 7 m/s reduces thermal contrast — flagged as a quality note if exceeded |
| Ambient temperature | Recorded at inspection start and end | Baseline for delta-T calculations in structural and earth wire components |
| Mean field temperature | Calculated from all module temperatures across the array | Reference for module anomaly delta-T classification |
If your report shows irradiance below 600 W/m2 during a significant portion of the inspection, request that Elion repeat those sections on a day with better conditions. We flag this proactively in our quality notes section.
Section 2: Reading the Anomaly Table
The anomaly table is the operational core of the report. Each row represents one finding.
| Column | What It Means | What to Do With It |
| Anomaly ID | Sequential identifier for this finding in this report | Use when communicating with Elion about a specific finding |
| Module ID or Component ID | Physical identifier in the plant layout map | Use to locate the module or component in your O&M system |
| GPS Coordinates | Latitude and longitude of the anomaly | Load into Google Maps or your GIS system to navigate to the exact location |
| Anomaly Type | Hotspot, bypass diode, PID, soiling, string fault, etc. | Determines which O&M action is appropriate — bypass diode requires module replacement, PID may require plant-level intervention |
| Delta-T | Temperature difference above mean field temperature for modules, or above ambient for other components | Key number for severity classification |
| Severity Class | Class 1, 2, or 3 | Class 3 = urgent action within weeks. Class 2 = plan for next O&M cycle. Class 1 = monitor at next inspection |
| Recommended Action | Specific recommendation from Elion’s engineers | First point of guidance for your O&M team |
Section 3: Thermal and RGB Image Pairs — How to Read Them
For each anomaly, the report includes two images: the thermal image showing the temperature pattern, and the RGB (visible) image of the same location. Always look at both.
- The thermal image tells you: what the temperature anomaly looks like, how large it is, and how it compares to surrounding modules
- The RGB image tells you: exactly which physical module has the anomaly and whether there is a visible cause such as soiling or damage
- A hotspot in thermal with a clearly visible soiling deposit in RGB is a cleaning issue, not a module defect
- A hotspot in thermal with a normal-looking module in RGB is a sub-surface issue — bypass diode, cell crack, or PID — requiring I-V curve tracing or EL imaging
Section 4: Annotated Aerial Map — How to Use It
The aerial map overlays all anomaly locations on a satellite or orthomosaic image of the plant. Use this map to:
- Identify spatial clusters — if Class 3 findings are concentrated in a specific section, there may be a systemic cause such as shading from a new structure or inverter problem
- Plan O&M access routes — the map shows which aisles to send maintenance personnel to for maximum efficiency
- Brief the maintenance team — a visual map is easier to brief from than a data table alone
Section 5: Priority Maintenance List — This is Your Action Plan
Each Class 3 finding is listed with the recommended action, component ID, and GPS location. Hand this directly to your O&M contractor. The list is sorted by severity so the most critical items are at the top.
What to Do After Receiving the Report
| Severity Class | Time to Act | Who to Involve | Action |
| Class 3 — Critical | Within 2 to 4 weeks | O&M contractor, plant manager | Expedite maintenance. For bypass diode failures, replace module or send for warranty claim. For Class 3 hotspots, disconnect the string and inspect at module level. |
| Class 2 — Significant | Within 3 to 6 months or next O&M visit | O&M contractor | Add to the next maintenance work order. Monitor generation data to confirm the string is still contributing. |
| Class 1 — Minor | At next annual inspection | O&M team — note in asset register | No immediate action. Include in next inspection scope to check if the anomaly has progressed. |
How to Cross-Reference the Report with SCADA Data
Every Class 3 and Class 2 finding should be cross-referenced with your plant SCADA generation data. Pull the production history for the string that includes the anomalous module. If the string has been underperforming relative to neighbouring strings for more than two weeks, this confirms the fault is active. If the string production is normal, the fault may be early-stage or intermittent — still worth addressing but at lower urgency.
Request an inspection and receive your IEC 62446-3 report
FAQ
Q: Can I dispute a finding in the report?
Yes. If you believe a finding is a thermal artefact rather than a true fault — for example, you believe a hot area is caused by a known temporary shading object — contact Elion with your evidence. We review the raw thermal data and RGB image for that finding and either confirm or revise the classification.
Q: What if the report shows fewer anomalies than expected?
A low anomaly count in a well-performing plant is a good result. If you expected more findings based on known SCADA underperformance, provide us the SCADA data for the underperforming strings before the next inspection. We will focus the analysis on those sections.