Systems and methods for data analytics for virtual energy audits and value capture assessment of buildings

Patent metadata

Field	Value
Publication number	US-2022383334-A1
Application number	US-202217815563-A
Country	US
Kind code	A1
Filing date	Jul 27, 2022
Priority date	May 10, 2018
Publication date	Dec 1, 2022
Grant date	—

Abstract

A system may provide virtual energy audits of one or more target buildings. The system may retrieve weather data and energy usage data specific to a given target building from a weather server and a utility server, respectively. The system may store predefined building characteristics corresponding to the given target building in local memory. Based on the weather data, energy usage data, and/or predefined building characteristics, the system may generate one or more building markers that characterize the energy usage and efficiency of the given target building. Building efficiency diagnostics and energy conservation prognostics may be generated based on the building markers and may be sent by the system to be displayed via a user interface of a client device. The energy conservation prognostics may include one or more energy conservation measure recommendations and corresponding predicted cost/energy savings.

First claim

1 - 27 . (canceled)

28 . A system for providing virtual energy audits comprising:
a processor; and at least one memory coupled with the processor, the at least one memory having software stored thereon which, when executed by the processor, causes the processor to:
retrieve input data corresponding to a target building, the input data comprising time-series energy usage data and building characteristic data, the time-series energy usage data comprising total power usage of the target building;
generate a plurality of building markers for the target building based on the input data,
automatically generate building efficiency diagnostics based on the plurality of building markers, wherein the building efficiency diagnostics include estimated on/off cycles of an HVAC system of the target building, the estimated on/off cycles of the HVAC system determined based on the time-series energy usage data; and
automatically send the building efficiency diagnostics to be displayed on a user interface.

29 . The system of claim 28 , wherein to retrieve input data, the software causes the processor to:
periodically retrieve new energy usage data corresponding to the target building from at least one of: a utility providing energy or a power sensor disposed at the target building; add the new energy usage data to the time-series energy usage data in the at least one memory; retrieve the time-series energy usage data corresponding to the target building; and retrieve the building characteristic data corresponding to the target building.

30 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate an effective building R-value marker by:
disaggregating the time-series energy usage data into a heating/cooling dataset and a load dataset; determining an interior heating load for a selected time period based on the load dataset, wherein the selected time period corresponds to a time period during which an interior temperature of the target building is substantially unchanging; determining an amount of energy being removed from an air-conditioned space of the target building based on the heating/cooling dataset for the selected time period; determining an exterior temperature of the target building for the selected time period based on weather data; estimating the interior temperature of the target building for the selected time period within a predetermined range; and generating the effective R-value building marker corresponding to a thermal insulation quality of the target building based on the amount of energy being removed, the interior heating load, the exterior temperature, and the interior temperature.

31 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a heating/cooling system turn on time building marker, a heating/cooling system turn off time building marker, and a heating/cooling system turn on wattage building marker by:
calculating a derivative of the time-series energy usage data to produce a derivative dataset defining changes in energy usage between timestamps of the time-series energy usage data; identifying heating/cooling system turn on times from the derivative dataset; identifying heating/cooling system turn off times from the derivative dataset; identifying a first mode of the identified heating/cooling system turn on times; identifying a second mode of the identified heating/cooling system turn off times; generating the heating/cooling system turn on wattage building marker based on observed energy usage changes occurring at the heating/cooling system turn on times; generating the heating/cooling system turn on time building marker equal to the first mode; and generating the heating/cooling system turn off times equal to the second mode.

32 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a diurnal pattern building marker by:
applying a chi-squared periodogram test to the time-series energy usage data to generate the diurnal pattern building marker.

33 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a rescheduling savings opportunity building marker by:
applying an analytical method to the time-series energy usage data to identify when the target building is unoccupied, wherein the analytical method is selected from at least one of: wavelet transform, two sample t-test, or paired t-test; determining that the HVAC system is active when the target building is unoccupied based on the time-series energy usage data; generating a recommendation to adjust a temperature setpoint of the HVAC system of the target building; and generating an estimated cost savings associated with adjusting the temperature setpoint according to the recommendation.

34 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate an energy usage change building marker by:
separating the time-series energy usage data into a plurality of single-year subsets; for each year represented in the plurality of single-year subsets, identifying significant change-points of the time-series energy usage data that occurred during a respective year; determining that a correlation between first and second significant change points is lower than a predetermined threshold; and flagging the first and second significant change points as corresponding to a retrofit time for the target building.

35 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a heating type building marker and a cooling type building marker by:
removing datapoints from the time-series energy usage data and weather data corresponding to holidays and weekends to produce modified energy usage data and modified weather data; applying a piecewise linear regression model to the modified energy usage data and time-series exterior temperature data of the modified weather data to produce a heating season trendline and a cooling season trendline; determining a first slope of the heating season trendline; determining a second slope of the cooling season trendline; comparing the first slope to a first predetermined threshold to determine a heating type of the target building; comparing the second slope to a second predetermined threshold to determine a cooling type of the target building; setting the heating type building marker equal to the determined heating type; and setting the cooling type building marker equal to the determined cooling type.

36 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a heating/cooling system size building marker by:
generating a heating/cooling system turn on time building marker; defining a subset of the time-series energy usage data as a set of datapoints corresponding to the heating/cooling system turn on time building marker; determining energy demand values for each of the set of datapoints; determining a mode of the energy demand values; and setting the heating/cooling system size building marker equal to the mode.

37 . The system of claim 28 , wherein to generate the plurality of building markers, the software causes the processor to generate a heating/cooling system oversized condition building marker by:
identifying a first subset of the time-series energy usage data corresponding to a heating season; identifying a second subset of the time-series ener

Assignees

• Univ Case Western Reserve

Inventors

• Abramson Alexis
• French Roger
• Pickering Ethan
• Hossain Mohammad
• Khalilnejad Arash
• Haddadian Rojiar

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06Q50/163

  Real estate management · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N20/20

  Ensemble learning · CPC title