This workflow is available on the KNIME Public Workflow Server:
This example shows how to predict a future value of a time series given its past values. The workflow shows how to remove seasonality from a time series and train an auto-regressive model for time series prediction. Key nodes for this use case are the Lag Column node, to provide past values and seasonality pattern. This example workflow works on time series of energy usage for smart meter clusters.
The full example is available as white paper in the white paper section.