Automated fault detection and diagnostics in a building management system

• US 9,606,520 B2
• Filed: 04/04/2012
• Issued: 03/28/2017
• Est. Priority Date: 06/22/2009
• Status: Active Grant

• Current Assignee: Johnson Controls Technology Company (Johnson Controls International plc)
• Original Assignee: Johnson Controls Technology Company (Johnson Controls International plc)
• Inventors: Noboa, Homero L., Drees, Kirk H., Wenzel, Michael
• Primary Examiner(s): Ali, Mohammad
• Assistant Examiner(s): PARK, JOHN C
• Application Number:
  US13/439,779
• Publication Number:
  US 20120259583A1
• Time in Patent Office: 1,819 Days
• Field of Search: 700/275
• US Class Current: 1/1
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G05B 2219/2642 Domotique, domestic, home control, automation, smart house
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
• CPC Class Codes: G06F 30/13 Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
• CPC Class Codes: H02J 13/00001 characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
• CPC Class Codes: H02J 13/00002 characterised by monitoring
• CPC Class Codes: H02J 13/00016 using a wired telecommunication network or a data transmission bus
• CPC Class Codes: Y02B 70/30 Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
• CPC Class Codes: Y02B 70/3225 Demand response systems, e.g. load shedding, peak shaving
• CPC Class Codes: Y02B 90/20 Smart grids as enabling technology in buildings sector smart grids supporting the management or operation of end-user stationary applications in general, or like technologies with no associated climate change mitigation effect Y04S20/00
• CPC Class Codes: Y02E 60/00 Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
• CPC Class Codes: Y04S 10/30 State monitoring, e.g. fault, temperature monitoring, insulator monitoring, corona discharge
• CPC Class Codes: Y04S 10/40 Display of information, e.g. of data or controls
• CPC Class Codes: Y04S 20/00 Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
• CPC Class Codes: Y04S 20/20 End-user application control systems
• CPC Class Codes: Y04S 20/221 General power management systems
• CPC Class Codes: Y04S 20/222 Demand response systems, e.g. load shedding, peak shaving
• CPC Class Codes: Y04S 40/124 using wired telecommunication networks or data transmission busses

Abstract

Systems and methods for automated fault detection in a building management system are shown and described. The systems and methods further involve organizing the data set into bins, each bin containing a plurality of the observed points for the variable of the building management system. The systems and methods further involve calculating, for each bin, a target parameter and an estimator of scale for the observed points. The systems and methods further involve detecting, for each bin, an outlier of the building management system by comparing a data point of the bin to the calculated target parameter and estimator of scale for the bin.

Claims

1. A computerized method for detecting outliers in a building management system having multiple operating modes, the method comprising:

receiving, at a building management system, a data set containing a plurality of observed points for a variable of the building management system, wherein each of the plurality of observed points is associated with a binning parameter indicating an operating mode of the building management system at a time when the observed point was generated, wherein the operating mode comprises high energy consumption mode and low energy consumption mode;

organizing the data set into multiple bins based on the binning parameters, each bin representing a subset of the data set and containing a plurality of the observed points for the variable of the building management system, wherein the observed points within each bin are related to each other by the binning parameters associated therewith, wherein each bin corresponds to a different operating mode of the building management system and contains the observed points that were generated when the building management system was operating in the corresponding operating mode;

calculating, for each bin, a target parameter and an estimator of scale for the observed points contained within the bin, each of the bins having a different target parameter and a different estimator of scale;

generating, for each bin, an upper control limit by adding a multiple of the calculated estimator of scale to the calculated target parameter for the bin and a lower control limit by subtracting a multiple of the calculated estimator of scale from the calculated target parameter for the bin, each of the bins having a different upper control limit and a different lower control limit;

detecting, for each bin, an outlier of the building management system by comparing a data point of the bin to the upper control limit and the lower control limit for the bin, wherein an outlier is detected if the data point is not within a bin-specific control limit interval between the upper control limit and the lower control limit, each of the bins having a different bin-specific control limit interval which characterizes operation of the building management system in the corresponding operating mode, wherein at least one of the bin-specific control limit intervals increases a likelihood of detecting the outlier relative to a control limit interval generated by performing the calculating and generating steps for an entirety of the data set;

outputting an indication of one or more of the outliers to at least one of a memory device, a user device, or another device on the building management system; and

wherein the variable of the building management system comprises a controllable variable and an uncontrollable variable;

wherein the controllable variable comprises at least one of occupancy of an area, space usage, and occupancy hours;

wherein the uncontrollable variable comprises at least one of outdoor air temperature, solar intensity, and degree days.

2. The method of claim 1, wherein the variable is an indicator of performance of a mechanical system of the building management system that maintains a setpoint.

3. The method of claim 2, wherein the variable is an exponentially weighted moving average of the error between an actual performance and an expected performance of the mechanical system.

4. The method of claim 1, wherein the binning parameters describe a condition or set of conditions that relate to performance changes of the variable.

5. The method of claim 1, wherein the binning parameter further indicates at least one of a time of day and an outside air temperature at the time when the observed point was generated.

6. The method of claim 1, further comprising causing a graphical user interface to be displayed, wherein the graphical user interface is configured to graphically represent the data set, target parameter, estimator of scale, upper control limit, and lower control limit.

7. The method of claim 1 further comprising:

detecting, for a monitored period, a fault of the building management system; and

causing an indication of the fault for the monitored period to be displayed by a user interface device.

8. The method of claim 7, wherein detecting a fault of the building management system for a monitored period comprises:

counting, for the monitored period, the number of outliers detected;

comparing the number of outliers detected during the monitored period to a threshold number of outliers under normal operating conditions; and

detecting a fault of the monitored period if the number of outliers detected is greater than the threshold number of outliers.

9. The method of claim 8, wherein the threshold number of outliers is determined by a statistical distribution.

10. The method of claim 9, wherein the statistical distribution is at least one of a binomial distribution, a normal distribution, or a Poisson distribution.

11. The method of claim 1, wherein the target parameter is at least one of a median or a mean, and wherein the estimator of scale is at least one of a robust standard deviation or a standard deviation.

12. The method of claim 2, wherein detecting an outlier does not require configuration information to identify operating modes of the mechanical system.

13. A controller for detecting outliers in a building management system having multiple operating modes, the controller comprising:

a processing circuit configured to receive a data set containing a plurality of observed points for a variable of the building management system, wherein each of the plurality of observed points is associated with a binning parameter indicating an operating mode of the building management system at a time when the observed point was generated, wherein the operating mode comprises high energy consumption mode and low energy consumption mode; and

wherein the processing circuit is further configured to organize the data set into multiple bins based on the binning parameters, each bin representing a subset of the data set and containing a plurality of the observed points for the variable of the building management system, wherein the observed points within each bin are related to each other by the binning parameters associated therewith, wherein each bin corresponds to a different operating mode of the building management system and contains the observed points that were generated when the building management system was operating in the corresponding operating mode;

wherein the processing circuit is further configured to calculate, for each bin, a target parameter and an estimator of scale for the observed points contained within the bin, each of the bins having a different target parameter and a different estimator of scale;

wherein the processing circuit is further configured to generate, for each bin, an upper control limit by adding a multiple of the calculated estimator of scale to the calculated target parameter for the bin and a lower control limit by subtracting a multiple of the calculated estimator of scale from the calculated target parameter for the bin, each of the bins having a different upper control limit and a different lower control limit;

wherein the processing circuit is further configured to detect, for each bin, an outlier of the building management system by comparing a data point of the bin to the upper control limit and the lower control limit for the bin, wherein an outlier is detected if the data point is not within a bin-specific control limit interval between the upper control limit and the lower control limit, each of the bins having a different bin-specific control limit interval which characterizes operation of the building management system in the corresponding operating mode, wherein at least one of the bin-specific control limit intervals increases a likelihood of detecting the outlier relative to a control limit interval generated by performing the calculating and generating steps for an entirety of the data set;

wherein the processing circuit is further configured to output an indication of one or more of the faults to at least one of a memory device, a user device, or another device on the building management system; and

wherein the variable of the building management system comprises a controllable variable and an uncontrollable variable;

wherein the controllable variable comprises at least one of occupancy of an area, space usage, and occupancy hours;

wherein the uncontrollable variable comprises at least one of outdoor air temperature, solar intensity, and degree days.

14. The controller of claim 13, wherein the variable is an indicator of performance of a mechanical system of the building management system that maintains a setpoint.

15. The controller of claim 14, wherein the variable is an exponentially weighted moving average of the error between an actual performance and an expected performance of the mechanical system.

16. The controller of claim 13, wherein the binning parameters describe a condition or set of conditions that relate to performance changes of the variable.

17. The controller of claim 13, wherein the binning parameter further indicates at least one of a time of day and an outside air temperature at the time when the observed point was generated.

18. The controller of claim 13, wherein the processing circuit is further configured to cause a graphical user interface to be displayed, the graphical user interface is configured to graphically represent the data set, target parameter, estimator of scale, upper control limit, and lower control limit.

19. The controller of claim 13, wherein the processing circuit is further configured to detect, for a monitored period, a fault of the building management system;

and wherein the processing circuit is further configured to cause an indication of the fault for the monitored period to be displayed by a user interface device.

20. The controller of claim 19, wherein detecting a fault of the building management system for a monitored period comprises:

counting, for the monitored period, the number of outliers detected; and

comparing the number of outliers detected during the monitored period to a threshold number of outliers under normal operating conditions; and

detecting a fault of the monitored period if the number of outliers detected is greater than the threshold number of outliers.

21. The controller of claim 20, wherein the threshold number of faults is determined by a statistical distribution.

22. The controller of claim 21, wherein the statistical distribution is at least one of a binomial distribution, a normal distribution, or a Poisson distribution.

23. The controller of claim 13, wherein the target parameter is at least one of a median or a mean, and wherein the estimator of scale is at least one of a robust standard deviation or a standard deviation.

24. The controller of claim 14, wherein detecting an outlier does not require configuration information to identify operating modes of the mechanical system.