Systems and methods for generating an energy usage model for a building

Patent metadata

Field	Value
Publication number	US-9575475-B2
Application number	US-201615016243-A
Country	US
Kind code	B2
Filing date	Feb 4, 2016
Priority date	Jun 22, 2009
Publication date	Feb 21, 2017
Grant date	Feb 21, 2017

Abstract

A building management system (BMS) includes a baseline model generator configured to receive an initial set of predictor variables for potential use in an energy usage model for a building, generate a first set of coefficients for the baseline energy usage model based on the initial set of predictor variables, remove one of the predictor variables from the initial set of predictor variables to create a subset of the initial set of predictor variables, generate a second set of coefficients for the baseline energy usage model based on the subset of the initial set of predictor variables, calculate a test statistic for the removed variable using a difference between the first set of coefficients and the second set of coefficients, and automatically select the removed predictor variable for use in the baseline energy usage model in response the test statistic exceeding a critical value.

First claim

What is claimed is:

1. A building management system comprising:
HVAC equipment configured to affect an environmental condition within a building by providing heating or cooling to the building;
one or more sensors configured to measure one or more predictor variables for potential use in a baseline energy usage model for the building, wherein the baseline energy usage model predicts an energy consumption of the HVAC equipment; and
a baseline model generator comprising a processing circuit configured to:
receive an initial set of predictor variables including the one or more measured predictor variables;
generate a first set of coefficients for the baseline energy usage model based on the initial set of predictor variables;
remove one of the predictor variables from the initial set of predictor variables to create a subset of the initial set of predictor variables;
generate a second set of coefficients for the baseline energy usage model based on the subset of the initial set of predictor variables;
calculate a test statistic for the removed variable using a difference between the first set of coefficients and the second set of coefficients; and
automatically select the removed predictor variable for use in the baseline energy usage model in response to the test statistic exceeding a critical value.

2. The system of claim 1 , wherein the processing circuit is configured to determine that the removed predictor variable is statistically significant to energy usage of the building in response to the test statistic exceeding the critical value.

3. The system of claim 1 , wherein the processing circuit is configured to:
generate the first set of coefficients by applying a regression analysis to the initial set of predictor variables; and
generate the second set of coefficients by applying a regression analysis to the subset of the initial set of predictor variables.

4. The system of claim 1 , wherein the initial set of predictor variables comprises a calculated predictor variable based on one or more of the measured predictor variables.

5. The system of claim 1 , wherein the processing circuit is configured to:
receive a user selection of one or more predictor variables via a user interface device; and
add one or more predictor variables to the initial set of variables based on the user selection.

6. The system of claim 1 , wherein the processing circuit is configured to:
receive a user selection of a data source via a user interface device;
identify one or more predictor variables provided by the data source; and
add the one or more identified predictor variables to the initial set of predictor variables.

7. The system of claim 1 , wherein the processing circuit is configured to:
collect data points for each of the predictor variables in the initial set of predictor variables, each data point having a data value and a time value;
identify a time period during which all of the predictor variables in the initial set of predictor variables have a threshold number of data points based on the time values for each of the data points; and
using the identified time period to generate the first set of coefficients for the baseline energy usage model.

8. The system of claim 1 , wherein the processing circuit is configured to:
use a first set of building data collected during a first time period to generate a third set of coefficients for the baseline energy usage model;
use a second set of building data collected during a second time period to generate a fourth set of coefficients for the baseline energy usage model; and
detect a fault in the building in response to a difference between the third set of coefficients and the fourth set of coefficients exceeding a threshold value.

9. The system of claim 1 , wherein the processing circuit is configured to:
use a first set of building data collected prior to implementing energy conservation measures to generate a third set of coefficients for the baseline energy usage model;
use a second set of building data collected after implementing the energy conservation measures to generate a fourth set of coefficients for the baseline energy usage model; and
estimate an amount of energy saving resulting from the energy conservation measures based on a difference between the third set of coefficients and the fourth set of coefficients.

10. The system of claim 9 , wherein the processing circuit is configured to:
apply the second set of building data to the baseline energy model with the third set of coefficients to estimate an energy usage of the building had the energy conservation measures not been implemented; and
compare the estimated energy usage to an actual energy usage of the building after implementing the energy conservation measures to estimate an amount of energy savings resulting from the energy conservation measures.

11. A method for generating a baseline energy usage model for a building, the method comprising:
operating HVAC equipment to affect an environmental condition within the building by providing heating or cooling to the building;
measuring one or more predictor variables for potential use in the baseline energy usage model using one or more sensors of the building, wherein the baseline energy usage model predicts an energy consumption of the HVAC equipment;
receiving, at a processing circuit, an initial set of predictor variables including the one or more measured predictor variables;
generating, by the processing circuit, a first set of coefficients for the baseline energy usage model based on the initial set of predictor variables;
removing, by the processing circuit, one of the predictor variables from the initial set of predictor variables to create a subset of the initial set of predictor variables;
generating, by the processing circuit, a second set of coefficients for the baseline energy usage model based on the subset of the initial set of predictor variables;
calculating, by the processing circuit, a test statistic for the removed predictor variable using a difference between the first set of coefficients and the second set of coefficients; and
automatically selecting, by the processing circuit, the removed predictor variable for use in the baseline energy usage model in response to the test statistic exceeding a critical value.

12. The method of claim 11 , further comprising determining that the removed predictor variable is statistically significant to energy usage of the building in response to the test statistic exceeding the critical value.

13. The method of claim 11 , wherein:
generating the first set of coefficients comprises applying a regression analysis to the initial set of predictor variables; and
generating the second set of coefficients comprises applying a regression analysis to the subset of the initial set of predictor variables.

14. The method of claim 11 , wherein the initial set of predictor variables comprises a calculated predictor variable based on one or more of the measured predictor variables.

15. The method of claim 11 , further comprising:
receiving a user selection of one or more predictor variables via a user interface device; and
adding one or more predictor variables to the initial set of predictor variables based on the user selection.

16. The method of claim 11 , further comprising:
receiving a user selection of a data source via a user interface device;
identifying one or more predictor variables provided by the data source; and
adding the one or more identified predictor variables to the initial set of variables.

17. The m

Assignees

• Johnson Controls Tech Co

Inventors

• Drees Kirk H
• Kummer James P
• Wenzel Michael J

Classifications

• H02J3/003 Primary

  Load forecast, e.g. methods or systems for forecasting future load demand · CPC title

• H02J2103/30

  Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks · CPC title

• H02J13/13

  characterised by the transmission of data to equipment in the power network · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• Y04S40/12

  characterised by data transport means between the monitoring, controlling or managing units and monitored, controlled or operated electrical equipment · CPC title