Virtual Sensor Modelling

Our Goal

We'd like to create a Virtual Sensor for a measurement we're currently taking by hand.

Plan

To do this, we will need a few steps.

1) Record the manual measurements, along with the precise time they were taken,
2) For each of those times, look up the machine state

Step 1a: Get Data

Collect your data and record it in a spreadsheet.

When you've collected a number of samples in various conditions, you can then export that information as a CSV file.

Ensure that your CSV file has a header row and one column is called either 'Time' or 'Date'.

Time, Vibration
2025-02-01 10:00:00, 21.4
2025-02-03 02:34:33, 0.5

Step 1b: Add an Upload Layer

To accept this CSV file for use in Capture, we add a get_upload layer.

{
	"type": "get_upload",
        "columns": ["Vibration"]
}

Step 2a: Capture the Machine State

Next, we get all of the context information we can about our machine.

{
	"type": "get_query",
	"query": "'Machine' ASSET AIPOINTS",
        "samples": 30,
	"comment": "Get Activity Details"
}

Step 2b: Capture the Middle Value

Finally, we'll get the 'middle' value out of the samples we picked up.

{
	"type": "flatten",
	"method": "middle",
        "comment": "Getting Sample Data"
}

Results

If you have a reasonable amount of context data, you'll have a huge amount of insight available to build an AI around your vibration.

Not only will this include the current state of your machine (ie. speed, current, voltage etc.), but also information about the entire production line, ambient conditions etc.

Virtual Sensor Modelling