With a Solar Inspection from Sitemark, you will get an overview of all solar anomalies present on your site. Each solar panel anomaly including thermal cause classification, visual cause classification, severity, temperature deltas and power production loss estimation.

Thermal Cause Classification

Based on the thermal data, Sitemark identifies the thermal cause of the anomaly. Sitemark can currently classify the following types of anomalies:

  • Hot Spot
  • Multi Hot Spot
  • Single Bypassed Substring
  • Double Bypassed Substring
  • Diode
  • Open Panel
  • Missing Panel
  • Potential Induced Degradation (PID)
  • Suspected Potential Induced Degradation (PID)
  • Open String
  • Junction Box

For more information about these thermal classifications, please read the this article.

Visual Cause Classification

Understanding the different causes of anomalies within your PV site will dramatically improve your ability to maintain and run an efficient solar park. Once you understand where and what is causing problems, it's much easier to take remedial action. Anomaly causes will tell you if you need to replace your panel or if you simply need to cut back vegetation. Sitemark can currently classify the following types of causes:

  • Soiling
  • Physical
  • Vegetation
  • Shadowing
  • Droppings

For more information about these visual classifications, please read the this article.

Loss Estimation

Sitemark has developed a loss estimation model that gives an indication of the possible energy yield loss that each anomaly is generating. The loss estimation helps you understand what regions of your site are underperforming. You can see how many anomalies you have per region. If you are using the same regions as your monitoring system, you can cross-check the MWh/year losses individually.

The loss estimation for a panel is calculated as follows:

Loss estimation = (% Panel loss * Panel Maximum Power) * Photovoltaic power potential

For this model to work, the following properties must be configured on your site:

  • Feed-in tariff
  • Photovoltaic power potential (PVOUT)
  • Panel Maximum Power (Pmax)

For help with regard to setting up site properties, please read the article.

Anomaly Severity

All anomalies detected are automatically classified into different severity levels. The severity evaluation helps in prioritizing which issues to address first on your site.


These are issues that require no immediate action, yet are essential to monitor over time.

  • Hot spot 0-5 °C DeltaT
  • Multi Hot spot 0-2,5 °C DeltaT
  • PID
  • Suspected PID
  • Junction Box


These are solar panels that show issues and are recommended to be replaced during the next maintenance cycle.

  • Hot Spot 5-15 °C DeltaT
  • Multi Hot Spot 2,5-7,5 °C DeltaT
  • Single Bypassed Substring


These are solar panels generating significant loss and are recommended to be acted on as soon as possible.

  • Hot Spot 15+ °C DeltaT
  • Multi Hot Spot 7,5+ °C DeltaT
  • Double Bypassed Substring
  • Open Panel
  • Open String
  • Missing Panel
  • Diode

Frequently Asked Questions

Is the anomaly localization and classification done using Artificial Intelligence (AI)?

Sitemark uses machine learning models to aid in the localization and classification of anomalies. To ensure the maximum possible accuracy, Sitemark deploys a HITL (human-in-the-loop) processing model. Machine learning detections are being verified by humans and errors are corrected and fed back into the model training process.

Did this answer your question?