Virtual sensors are software-based tools that simulate physical sensors, often by processing existing sensor data or other relevant information. We can implement a virtual sensor by creating a machine learning model. This model extracts information from other measurements and estimates the quantity of interest. Virtual sensing techniques are helpful to provide economic and feasible alternatives to impractical or costly physical measurements. This approach promises cost savings because the company virtually performs physical, costly tasks.


  1. Objectives.
  2. Benefits.
  3. Approach.
  4. Conclusions.


Virtual sensing tries to model our target variable, so it is unnecessary to measure it physically.

The challenge is to get the correct variables for our model to predict the quantity of interest.


It allows obtaining measures using a machine learning model, reducing costs and work hours, as these measures must usually be obtained in a laboratory.




Virtual sensing is to create an approximation model that takes in some input variables related to our target and estimates the latter.

Neural networks can model the correct values of the target variable, so we don’t need to take physical measures.

That allows saving costs and time on measure’s obtention.


Virtual sensors save companies money and time since measure-taking becomes a software issue, improving planning and decision-making. Neural Designer uses machine learning to build predictive models representing various variables associated with virtual sensing.

Related posts: