Building control system with peer analysis based on weighted outlier detection

1. A system for assessing relative performance among a plurality of heating, ventilation, or air condition (HVAC) assets, the system comprising a processing circuit configured to:

identify a peer group comprising two or more of the plurality of HVAC assets having a common characteristic, the plurality of HVAC assets operable to affect one or more environmental conditions of one or more buildings; generate values of a peer metric for the HVAC assets in the peer group based on corresponding operation data associated with the HVAC assets;

perform a weighted outlier detection process using the values of the peer metric, wherein the weighted outlier detection process comprises assigning weights to the values of the peer metric, the weights having a positive correlation with corresponding amounts of the operation data used to generate the values of the peer metric; and

initiate an automated action in response to detecting an outlier HVAC asset, wherein the automated action comprises adjusting the operation data of at least one of the outlier HVAC asset or one or more other HVAC assets of the peer group based on a result of the weighted outlier detection process.

2. The system of claim 1 , wherein the peer metric comprises at least one peer metric selected from a plurality of peer metrics based on the peer group.

3. The system of claim 1 , wherein:

the operation data comprise a measured temperature and a setpoint temperature for the plurality of HVAC assets; and

the peer metric comprises a temperature control error based on the measured temperature and the setpoint temperature for each of the plurality of HVAC assets of the peer group.

4. The system of claim 1 , wherein:

the operation data comprise an actuation signal for a flow control device for the plurality of HVAC assets; and

the peer metric comprises a control effort based on the actuation signal for each of the plurality of HVAC assets of the peer group.

5. The system of claim 1 , wherein:

the operation data comprise a compressor suction pressure and a compressor discharge pressure for the plurality of HVAC assets; and

the peer metric comprises a compression ratio, based on the compressor suction pressure and the compressor discharge pressure, for each of the plurality of HVAC assets of the peer group.

6. The system of claim 1 , the processing circuit further configured to:

generate values of a plurality of peer metrics for the HVAC assets in the peer group based on corresponding operation data associated with the HVAC assets, wherein the operation data comprise a number of off times less than or equal to a first time threshold, a number of cycles, and a number of cycle times less than or equal to a second time threshold for the plurality of HVAC assets; and

the plurality of peer metrics comprises an off time metric and a cycle time metric based on the number of off times less than or equal to a first time period, number of cycles, and number of cycle times less than or equal to a second time period for each of the plurality of HVAC assets of the peer group.

7. The system of claim 1 , wherein the plurality of HVAC assets comprises a plurality of sensors configured to obtain the operation data associated with the plurality of HVAC assets.

8. The system of claim 1 , wherein:

the weighted outlier detection process is a univariate process comparing the peer metric across the plurality of HVAC assets for the peer group; and

the univariate process comprises a weighted Generalized Extreme Studentized Deviate test.

9. The system of claim 1 , wherein:

the weighted outlier detection process is a multivariate process comparing a plurality of peer metrics for the plurality of HVAC assets in the peer group; and

the multivariate process comprises a modified Sequential Application of Wilks's Multivariate Outlier Test.

10. The system of claim 1 , the outlier detection process comprising:

calculating a mean vector for a set comprising values of a plurality of peer metrics;

determining a matrix of the sums of squares and cross-products (“SSCP matrix”) for deviation scores of the set of values using the mean vector;

weighting the SSCP matrix, wherein the weighting is a function of an amount of operation data used to generate the values of the plurality of peer metrics;

determining a potential outlier HVAC asset from the plurality of HVAC assets as the HVAC asset corresponding to the peer metric value whose removal leads to the greatest reduction in the determinant of the weighted SSCP matrix;

calculating a weighted Wilk's Statistic using the peer metrics associated with the potential outlier HVAC asset and the weighted SSCP matrix; and

comparing the weighted Wilk's Statistic to a critical value to determine the outlier HVAC asset.

11. A method for assessing relative performance among a plurality heating, ventilation, or air conditioning (HVAC) assets, the method comprising:

identifying a peer group comprising two or more of the plurality of HVAC assets having a common characteristic, the plurality of HVAC assets operable to affect one or more environmental conditions of one or more buildings; generating values of a peer metric for the HVAC assets in the peer group based on corresponding operation data associated with the HVAC assets;

performing a weighted outlier detection process using the values of the peer metric, wherein the weighted outlier detection process comprises assigning weights to the values of the peer metric, the weights having a positive correlation with corresponding amounts of the operation data used to generate the values of the peer metric; and

initiating an automated action in response to detecting an outlier HVAC asset, wherein the automated action comprises adjusting the operation data of at least one of the outlier HVAC assets or one or more other HVAC assets of the peer group based on a result of the weighted outlier detection process.

12. The method of claim 11 , wherein the automated action further comprises displaying the results of the weighted outlier detection process to a user.

13. The method of claim 11 , further comprising identifying a first time period for wherein the operation data associated with the plurality of HVAC assets of the peer group belongs to the first time period.

14. The method of claim 11 , wherein:

the operation data comprise a measured temperature and a setpoint temperature for the plurality of HVAC assets; and

the peer metric comprises a temperature control error, based on the measured temperature and the setpoint temperature, for each of the plurality of HVAC assets of the peer group.

15. The method of claim 11 , wherein:

the operation data comprise an actuation signal for a flow control device for the plurality of HVAC assets; and

the peer metric comprises a control effort based on the actuation signal for each of the plurality of HVAC assets of the peer group.

16. The method of claim 11 , wherein:

the operation data comprise a compressor suction pressure and a compressor discharge pressure for the plurality of HVAC assets; and

the peer metric comprises a compression ratio, based on the compressor suction pressure and the compressor discharge pressure, for each of the plurality of HVAC assets of the peer group.

17. The method of claim 11 , further comprising:

generating values of a plurality of peer metrics for the HVAC assets in the peer group based on corresponding operation data associated with the HVAC assets; wherein the operation data comprise a number of off times less than or equal to a first time period, a number of cycles, and a number of cycle times less than or equal to a second time period for the plurality of HVAC assets; and

the plurality of peer metrics comprise an off time metric and a cycle time metric based on the number of off times less than or equal to a first time threshold, number of cycles, and number of cycle times less than or equal to a second time threshold for each of the plurality of HVAC assets of the peer group.

18. The method of claim 11 , wherein:

the operation data comprise an energy consumption and heating or cooling output for the plurality of HVAC assets; and

the peer metric comprises an efficiency based on the energy consumption and the heating or cooling output for each of the plurality of HVAC assets of the peer group.

19. The method of claim 11 , wherein:

the weighted outlier detection process is at least one of:

a univariate process comparing the peer metric across the plurality of HVAC assets of the peer group; and a

multivariate process comparing a plurality of peer metrics for the plurality of HVAC assets in the peer group of a plurality of types.