Correcting variable for bias


#1

Hi, I have 2 humidity sensors. Before deploying them in the field I’m checking them for consistency by reading them right next to one another. Both work, and track the humidity, but one of them is higher than the other. So I want to correct for that. I added a manual variable (humidity-correction) to enter the bias correction (and added it as a manual input to the dashboard, and put my bias correction number -119 in, then added synthetic variable (humidity-difference) to show the corrected historical data, but I get no result in the dashboard using synthetic expression (which validates with a ‘tick’):

humidity_out-humidity_in+fill_missing( humidity-correction)

When I use a constant instead it works:

humidity_out-humidity_in-119

What am I doing wrong?


#2

Hi there, to perform math operations between two variables, their dots must have the same timestamp. If your variables’ sample time is different, you should use the fill_missing() function as you have tried, but this function has as input raw time series rather than constants:

fill_missing({{raw_1}} + {{raw_2}})

You should set as arguments to the function your time series. For more information, please refer here.

All the best