Model deviation
Model deviation
| Available on |
  
|
Model deviation shows the average difference between the sensor values and the network simulation values. The smaller the model deviation, the closer the real data and the simulated values are. The goal is for the model deviation to be as small as possible, with the minimum value being 0%.
How to visualize model deviation insights
Qatium checks the model deviation every hour. To visualize the deviation values, open the Model deviation panel 
The assets with sensors will be highlighted with a halo in your map in:
- Green: when the deviation is below 10%
- Orange: when between 10% and 20%
- Red: when the deviation is above 20%
The panel uses the same color code to display information about:
- The overall deviation for flow, pressure, and tank volume values.
- The average error per each value.
This information is useful to locate the assets where there’s a higher deviation. So you can determine if something unplanned is happening in the network (like a pipe burst), hence where you might need to focus your efforts.
💡 Tip from Q: If you prefer, I can interact with the deviation panel for you. Just ask me about the network average error; or about flow, pressure and tank volume in particular. I’ll give you all the details and fly you to the most deviated elements.
⚠️ Important: We calculate accuracy and deviation metrics only within the base version of the network (My network). In the scenario version (My scenario), Qatium will neither update the accuracy of calculations nor display the deviations.