Energy disaggregation methods and systems using weather data
The system may have an energy load data interface to receive energy load data originating from energy use sensors for the building; a weather data interface to receive weather data for a location that includes the building; a weather adjustment pre-processor to process the energy load data and the weather data and to determine a weather-dependent energy use component of the energy load data and a weather-independent energy use component of the energy load data; a baseline adjustment pre-processor to process the weather-independent energy use component of the energy load data and determine a baseline energy use component of the energy load data, wherein the baseline adjustment pre-processor is configured to remove the baseline energy use component from the weather-independent energy use component to determine a variable energy use component of the energy load data; and an energy use disaggregator to process the variable energy use component of the energy load data and determine a plurality of time varying load components of the energy load data.
1. An energy disaggregation system for a building, comprising:
an energy load data interface to receive energy load data originating from energy use sensors for the building;
a weather data interface to receive weather data for a location that includes the building;
a weather adjustment pre-processor to process the energy load data and the weather data and to determine a weather-dependent energy use component of the energy load data and a weather-independent energy use component of the energy load data;
a baseline adjustment pre-processor to process the weather-independent energy use component of the energy load data and determine a baseline energy use component of the energy load data, wherein the baseline adjustment pre-processor is configured to remove the baseline energy use component from the weather-independent energy use component to determine a variable energy use component of the energy load data; and
an energy use disaggregator to process the variable energy use component of the energy load data and determine a plurality of time-varying load components of the energy load data;
the system further comprising:
an occupancy engine to estimate periods of occupancy for the building using the energy load data wherein the energy use disaggregator uses the estimated periods of occupancy to determine the plurality of time-varying load components of the energy load data;
a data partition engine to apply a clustering model to the energy load data to determine one or more partitions within the energy load data; and
a temporal processing engine to segment and aggregate the energy load data over a set of predefined time periods, wherein the temporal processing engine is independently applied to the partitions determined by the data partition engine,
wherein the output of the temporal processing engine is used by the occupancy engine to estimate periods of occupancy for the building;
the system further configured to:
i) determine a set of control actions available within the building that relate to one or more of the plurality of time-varying load components;
ii) predict a change in said one or more of the plurality of time-varying load components in response to one or more of the set of control actions;
iii) output recommendations for control actions that are predicted to reduce the said one or more of the plurality of time-varying load components;
iv) receive a selection of one or more control actions within the recommendations for control actions; and
v) implement the selected one or more control actions within the building;
the one or more control actions comprising at least one of the following:
a) in relation to one or more time-varying load components, switching off or changing one or more operating parameters for one or more appliances and/or items of equipment within the building;
b) in relation to a small power load component, turning off or turning down one or more appliances or items of equipment within the building;
c) in relation to a lighting load component, turning off or turning down a set of lights;
d) in relation to a heating, ventilation and air conditioning (HVAC) load component, changing a desired internal temperature, and/or opening and/or closing ventilation conduits.
2. The system according to claim 1 , wherein the energy load data comprises time-series energy consumption measurements for the building and wherein the components of the energy load data are determined for each of a plurality of time samples.
3. The system according to claim 1 , wherein the baseline energy use component is determined as a predefined portion of the weather-independent energy use component of the energy load data.
4. The system according to claim 1 , wherein the time-varying load components comprise one or more of:
a lighting load component;
a heating, ventilation and air conditioning (HVAC) load component;
a small power load component; and
a miscellaneous load component.
5. The system according to claim 1 , wherein the time-varying load components comprises a lighting load component and system further comprises:
a lighting data interface to receive lighting operating data for the building,
wherein the energy use disaggregator determines the lighting load component based on one or more of the lighting operating data and a lighting power density.
6. The system according to claim 1 , wherein the time-varying load components comprises a small power load component and the energy use disaggregator determines the small power load component based on one or more of the estimated periods of occupancy and a small power density.
7. The system according to claim 1 , wherein the weather adjustment pre-processor uses the estimated periods of occupancy to fit separate linear regressions to the energy load data for occupied and un-occupied periods within a day.
8. A method for disaggregating energy use for a building, the method comprising:
obtaining energy load data originating from energy use sensors for the building;
obtaining weather data for a location that includes the building;
processing the energy load data and the weather data to determine a weather-dependent energy use component of the energy load data and a weather-independent energy use component of the energy load data;
processing the weather-independent energy use component of the energy load data to determine a baseline energy use component of the energy load data;
removing the baseline energy use component from the weather-independent energy use component to determine a variable energy use component of the energy load data;
processing the variable energy use component of the energy load data to determine a plurality of time-varying load components of the energy load data; and
outputting visualization data indicating at least the weather-dependent energy use component, the baseline energy use component and the plurality of time-varying load components;
estimating periods of occupancy for the building using the energy load data; and using the estimated periods of occupancy to determine the plurality of time-varying load components of the energy load data;
applying a clustering model to the energy load data to determine one or more partitions within the energy load data;
temporally segmenting the energy load data over a set of predefined time periods on a partition basis;
aggregating the temporally segmented energy load data on a partition basis; and
separately determining estimated periods of occupancy for the building for each partition;
determining a set of control actions available within the building that relate to one or more of the plurality of time-varying load components;
predicting a change in said one or more of the plurality of time-varying load components in response to one or more of the set of control actions;
outputting recommendations for control actions that are predicted to reduce the said one or more of the plurality of time-varying load components;
receiving a selection of one or more control actions within the recommendations for control actions; and
implementing the selected one or more control actions within the building,
the one or more control actions comprising at least one of the following:
a) in relation to one or more time-varying load components, switching off or changing one or more operating parameters for one or more appliances and/or items of equipment within the building;
b) in relation to a small power load component, turning off or turning down one or more appliances or items of equipment within the building;
c) in relation to a lighting load component, turning off or turning down a set of lights;
d) in relation to a heating, ventilation and air conditioning (HVAC) load component, changing a desired internal temperature, and/or opening and/or closing ventilation conduits.
9. The method according to claim 8 , the energy load data and the weather data comprise time-series measurements and wherein the visualisation data comprises a plurality of time-series samples, the method further comprising displaying the visualisation data on a display device.
10. The method according to claim 8 , wherein the baseline energy use component is determined as a predefined portion of the weather-independent energy use component of the energy load data.
11. The method according to claim 8 , wherein the time-varying load components comprise one or more of:
a lighting load component;
a heating, ventilation and air conditioning (HVAC) load component;
a small power load component; and
a miscellaneous load component.
12. The method according to claim 8 , wherein the time-varying load components comprises a lighting load component and the method comprises:
obtaining lighting operating data for the building; and
determining the lighting load component based on one or more of the lighting operating data and a lighting power density.
13. The method according to claim 8 , wherein the time-varying load components comprises a small power load component and the method comprises:
determining the small power load component based on one or more of the estimated periods of occupancy and a small power density.
14. The method according to claim 8 , comprising:
using the estimated periods of occupancy to fit separate linear regressions to the energy load data for occupied and un-occupied periods within a day.
15. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 8 .