Icon

03_​ETL_​Energy_​autocorr_​stats

Usage Measures vs. auto-correlation on Energy Consumption Time Series

The theme ingredient in this workflow is Energy Consumption Time Series. What kind of variables can we extract from energy consumption data? Here we work on:
1. Usage Measures - average and in % for weekdays and day times for each meter ID time series
2. Auto-correlation for single selected meter ID time series to detect seasonality

This workflow focuses on the variables we can extract from energy consumption data.For more information see the workflow metadata. Find it here: View -> Description Usage MeasuresThe energy used is calculated for each meter ID in average and as percentagefor: - day times (morning, evening, afternoon, etc...) - week days (Monday, Tuesday, etc ... and business days vs. week ends) Auto-correlation MatrixHere we calculate the auto-correlation matrix for a single selectedmeter ID. Auto-correlation is calculated on 100 past samples.This number can be changed in the "Normalize & Lag" metanode. year, month,day of week,week of yearhour,minutesconvert proprietary date formatinto datetime valuesintra-week kW % usageby meter IDintra-day kW % usageby meter IDread only 1 of 6 filesand only 500K rowsdate_houras RowIDtotal KW per houron date & timevs. meter IDselectjust onemeter ID Date FieldExtractor (legacy) Daily Values Time FieldExtractor (legacy) String to datetime Week Day (%) Intra-daysegments (%) Hourly Values Joiner File Reader Line Plot (local) RowID Pivoting Find Seasonality Select Meter ID Normalize & Lag Linear Correlation This workflow focuses on the variables we can extract from energy consumption data.For more information see the workflow metadata. Find it here: View -> Description Usage MeasuresThe energy used is calculated for each meter ID in average and as percentagefor: - day times (morning, evening, afternoon, etc...) - week days (Monday, Tuesday, etc ... and business days vs. week ends) Auto-correlation MatrixHere we calculate the auto-correlation matrix for a single selectedmeter ID. Auto-correlation is calculated on 100 past samples.This number can be changed in the "Normalize & Lag" metanode. year, month,day of week,week of yearhour,minutesconvert proprietary date formatinto datetime valuesintra-week kW % usageby meter IDintra-day kW % usageby meter IDread only 1 of 6 filesand only 500K rowsdate_houras RowIDtotal KW per houron date & timevs. meter IDselectjust onemeter ID Date FieldExtractor (legacy) Daily Values Time FieldExtractor (legacy) String to datetime Week Day (%) Intra-daysegments (%) Hourly Values Joiner File Reader Line Plot (local) RowID Pivoting Find Seasonality Select Meter ID Normalize & Lag Linear Correlation

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
On NodePit since: 2025-02-18
Last update: 2025-05-12
Created with KNIME version: v4.1.1
Tags: ETLKNIMEdata manipulationtime seriesenergy consumptionauto-correlationcorrelationtime analysistime series analysis

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!