SYSTEM FOR MULTIDIMENSIONAL DATA-DRIVEN UTILITY BASELINING

• US 20090216469A1
• Filed: 02/27/2008
• Published: 08/27/2009
• Est. Priority Date: 02/27/2008
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Rieger, Josef, Marik, Karel
• Granted Patent: US 8,200,456 B2
• Time in Patent Office: Days
• US Class Current: 702/61
• CPC Class Codes: G05B 15/02 electric

Abstract

A system for utility base lining records historic values of utility loads for regions within a facility. The system also records historic values of independent variables such as outside temperature, time, date, workday versus non-workday, and occupancy. A similar data selector seeks out similar times in the past and submits the data from those times to a base line estimator which produces a baseline estimate. Differences between the current load and the estimated load can trigger alarms or investigations to determine why the utility load has changed. False alarms can occur when the utility load drifts over time. Detecting drift can help reduce false alarms or help in the investigation of alarms. A graphical user interface (GUI) can present an easily understood representation of the regions, load estimation errors, alarms, and detected drift.

Claims

1. A system comprising:

at least one sensor wherein each sensor produces a load measurement indicating the amount of energy currently consumed in one of at least one regions;

a means for determining independent variable values wherein the independent variables comprise outdoor temperature, date, time of day, and region;

a plurality of independent variable measurements indicating the values of the independent variables at different times;

a plurality of load measurements indicating the values of the sensed load measurements at different times;

a database storing the load measurements and the independent variable measurements as historical data;

a similar data selector that accepts a recent measurement and produces similar historical data from the historical data wherein the similar historical data comprises load measurements and independent variable measurements taken when the independent variable measurements were similar to the recent measurement; and

a baseline estimation module that accepts the historical load data set and produces at least one baseline estimate for the recent measurement.

2. The system of claim 1 further comprising a trigger that triggers the production of the similar historical data set and the at least one baseline estimate.

3. The system of claim 1 wherein the independent variables further comprise workday versus non-workday.

4. The system of claim 1 wherein the independent variables further comprise occupancy.

5. The system of claim 1 wherein the baseline estimation module comprises a linear regression model.

6. The system of claim 1 further comprising a graphical user interface wherein the recent measurement is a current independent variable measurement such that the at least one baseline estimate is at least one current baseline estimate;

and wherein the graphical user interface presents the at least one region with an indication of the difference between the current load measurement and the current baseline estimate for the at least one region.

7. A system comprising:

at least one sensor wherein each sensor produces a sensed load measurement indicating the amount of energy currently consumed in one of at least one regions;

a means for determining independent variable values wherein the independent variables comprise outdoor temperature, date, time of day, and region;

a plurality of independent variable measurements indicating the values of the independent variables at different times;

a plurality of load measurements indicating the values of the sensed load measurements at different times;

a database storing the load measurements and the independent variable measurements as historical data;

a similar data selector that accepts a recent measurement and a reference period and produces similar reference period data from the historical data;

a baseline estimation module that accepts the similar reference period data and produces at least one reference period baseline estimate for the recent measurement; and

a drift detection module that identifies baseline drift by comparing the reference period baseline estimates to the load measurements.

8. The system of claim 7 further comprising at least one drift detection rule wherein the drift detection module is a rule based drift detection module.

9. The system of claim 7 wherein the independent variables further comprise workday versus non-workday.

10. The system of claim 7 wherein the independent variables further comprise occupancy.

11. The system of claim 7 wherein the similar data selector is a distance based similar data selector.

12. The system of claim 7 wherein the baseline estimation module comprises a linear regression model.

13. The system of claim 7 further comprising a graphical user interface that presents the at least one region with an indication of the difference between the load measurements and the baseline estimates for the at least one region.

14. The system of claim 7 further comprising a graphical user interface that indicates the presence of baseline drift in any of the at least one region.

15. The system of claim 7 further comprising:

at least one drift detection rule wherein the drift detection module is a rule based drift detection module;

at least one data selection rule wherein the similar data selector is a rule based similar data selector; and

a graphical user interface that presents the at least one region with an indication of the difference between the load measurements and the baseline estimates for the at least one region and wherein the graphical user interface also indicates the presence of baseline drift in any of the at least one region;

wherein the independent variables further comprise workday versus non-workday and building occupancy; and

wherein the baseline estimation module comprises a linear regression model.

16. A system comprising:

a means for obtaining a plurality of load measurements for at least one region and covering a plurality of time periods for each region wherein each load measurement is an indication of the amount of energy consumed in one region over a known time period;

a means for storing the load measurements;

a means for obtaining a plurality of independent variable measurements corresponding to the values of independent variables at known times wherein the independent variables comprise outdoor temperature, date, and time of day;

a means for storing historical data comprising the load measurements and independent variable measurements; and

a means for producing at least one baseline estimate.

17. The system of claim 16 further comprising a means for determining drift.

18. The system of claim 17 further comprising a means for displaying the drift.

19. The system of claim 16 further comprising a means for displaying the difference between the load measurements and the baseline estimates.