Normalization
The Normalization processor scales incoming values into a 0 to 1 range, using a window of values to calculate the scale.
Why use Normalization
Use this processor when downstream calculations or models expect values on a consistent 0-to-1 scale rather than each tag's raw units, for example, normalizing sensor readings that have different native ranges before feeding them into a shared model.
How it works
The processor first observes a window of values and scales the next value between 0 and 1, based on the normalization method you choose in Mode. After that initial window, it continues as a moving window: each subsequent value is scaled between 0 and 1 relative to the window immediately before it.
Parameters
Parameter | Details |
|---|---|
Mode | The normalization method to apply: Min-Max scaling or Average Standard Deviation. |
Window Size | The number of values to observe when calculating each window's scale. |
Pass Through Value | Determines whether the processor's output is the normalized value or the original unchanged input value. |