System and method for automatic detection and clustering of variable air volume units in a building management system

1. A building management system comprising:

one or more memory devices configured to store instructions thereon that, when executed by one or more processors, cause the one or more processors to:

receive one or more of variable air volume unit (VAV) data or air handling unit (AHU) data from one or more data sources;

determine that one or more VAVs qualify as a suspect VAV in response to one or more conditions being satisfied by at least one of (i) the VAV data or (ii) the VAV data and the AHU data;

classify the suspect VAV as a first type of suspect VAV in response to a first set of the one or more conditions being satisfied;

classify the suspect VAV as a second type of suspect VAV in response to a second set of the one or more conditions being satisfied;

in response to the suspect VAV being classified as the first type of suspect VAV, remove a portion of the VAV data associated with the suspect VAV from a calculation of a duct static pressure setpoint (DPT-SP) for an AHU and operate building equipment to affect an airflow provided to one or more building zones using an adjusted value of the DPT-SP after removing the portion of the VAV data associated with the suspect VAV from the calculation of the DPT-SP; and

in response to the suspect VAV being classified as the second type of suspect VAV, increase or decrease the value of the DPT-SP and operate the building equipment to affect the airflow provided to the one or more building zones using the increased or decreased value of the DPT-SP.

2. The building management system of claim 1 , wherein the suspect VAV is identified as a rogue VAV if the first set of the one or more conditions being satisfied, an undersized VAV if the second set of the one or more conditions being satisfied, or an oversized VAV if a third set of the one or more conditions being satisfied.

3. The building management system of claim 1 , wherein the instructions further cause the one or more processors to tag the suspect VAV with one or more of a mechanical failure, a flow station that measures the air velocity pressure has failed, or an overridden damper.

4. The building management system of claim 3 , wherein the instructions further cause the one or more processors to employ a machine learning model for tagging the suspect VAV.

5. The building management system of claim 1 , wherein the AHU data includes one or more of a supply fan status (SF-S) or a duct static pressure (DPT), and the VAV data includes one or more of a discharge air velocity pressure (DA-VP), a supply air flow (SA-F), a supply air flow setpoint (SAF-SP), a damper output (DPR-O), a zone temperature (ZN-T), or a zone temperature setpoint (ZNT-SP).

6. The building management system of claim 5 , wherein the suspect VAV is tagged as a rogue VAV if the AHU is on, a VAV damper position is set to a maximum, and a percentage error between the DPT and the DPT-SP is less than a first predetermined percentage.

7. The building management system of claim 6 , wherein the suspect VAV is further tagged as having a loose damper or mechanical failure if the DPR-O is greater than a second predetermined percentage and the SA-F is less than SAF-SP,

wherein the suspect VAV is further tagged as a faulty flow station connected to the suspect VAV if the DA-VP has a constant value over a first predetermined amount of time; and

wherein the suspect VAV is further tagged as having an overridden damper if the DPR-O is greater than ZN-T, and the SA-F is greater than the SAF-SP.

8. The building management system of claim 5 , wherein the suspect VAV is tagged as oversized if the AHU is on, a percentage difference between the ZN-T and the ZNT-SP is less than a third predetermined percentage, a difference between SA-F and SAF-SP is less than a fourth predetermined percentage, and the SAF-SP is at a minimum limit.

9. The building management system of claim 8 , wherein the suspect VAV is further tagged as having an overridden damper if the DPR-O has a constant value for 2 or more consecutive operation periods.

10. The building management system of claim 5 , wherein the suspect VAV is tagged as undersized if the AHU is on, the ZN-T is greater than the ZNT-SP, VAV damper output is less than 100 percent, a difference between the SA-F and the SAF-SP is within a predetermined threshold, and the SAF-SP is at a maximum limit.

11. The building management system of claim 10 , wherein the instructions further cause the one or more processors to cluster two or more of the VAVs that are tagged as undersized VAVs together.

12. A method of monitoring and controlling building equipment, the method comprising:

receiving one or more of variable air volume unit (VAV) data or air handling unit (AHU) data from one or more data sources;

determining that one or more VAVs qualifies as a suspect VAV in response to one or more conditions being satisfied by at least one of (i) the VAV data or (ii) the VAV data and the AHU data;

classifying the suspect VAV as a first type of suspect VAV in response to a first set of the one or more conditions being satisfied;

classifying the suspect VAV as a second type of suspect VAV in response to a second set of the one or more conditions being satisfied;

in response to the suspect VAV being classified as the first type of suspect VAV, removing a portion of the VAV data associated with the suspect VAV from a calculation of a duct static pressure setpoint (DPT-SP) for an AHU and operating building equipment to affect an airflow provided to one or more building zones using an adjusted value of the DPT-SP after removing the portion of the VAV data associated with the suspect VAV from the calculation of the DPT-SP; and

in response to the suspect VAV being classified as the second type of suspect VAV, increasing or decreasing the value of the DPT-SP and operating the building equipment to affect the airflow provided to the one or more building zones using the increased or decreased value of the DPT-SP.

13. The method of claim 12 , wherein the determining the one or more VAVs as the suspect VAV includes tagging the suspect VAV as:

a rogue VAV if the first set of the one or more conditions being satisfied with one or more of having a mechanical failure, a flow station that measures the air velocity pressure has failed, or an overridden damper;

an undersized VAV if the second set of the one or more conditions being satisfied with one or more of having an overridden damper or having flow limits that require verification; and

an oversized VAV if a third set of the one or more conditions being satisfied with one or more of having an overridden damper or having flow limits that exceed a predetermined value.

14. The method of claim 13 , wherein the AHU data includes one or more of a supply fan status (SF-S) or a duct static pressure (DPT), and the VAV data includes one or more of a discharge air velocity pressure (DA-VP), a supply air flow (SA-F), a supply air flow setpoint (SAF-SP), a damper output (DPR-O), a zone temperature (ZN-T), or a zone temperature setpoint (ZNT-SP).

15. The method of claim 14 , wherein the suspect VAV is further tagged as:

the rogue VAV if the AHU is on, a VAV damper position is set to a maximum, and a percentage error between the DPT and the DPT-SP is less than a first predetermined percentage;

having a loose damper or mechanical failure if the DPR-O is greater than a second predetermined percentage and the SA-F is less than SAF-SP,

a faulty flow station connected to the suspect VAV if the DA-VP has a constant value over a first predetermined amount of time; and

as having an overridden damper if the DPR-O is greater than ZN-T, and the SA-F is greater than the SAF-SP.

16. The method of claim 14 , wherein the suspect VAV is further tagged as:

oversized if the AHU is on, a percentage difference between the ZN-T and the ZNT-SP is less than a third predetermined percentage, a difference between SA-F and SAF-SP is less than a fourth predetermined percentage, and a supply air flow setpoint is at a minimum limit; and

having an overridden damper if the DPR-O has a constant value for 2 or more consecutive operation periods.

17. The method of claim 14 , wherein the suspect VAV is further tagged as undersized if the AHU is on, the ZN-T is greater than the ZNT-SP, VAV damper output is less than 100 percent, a difference between the SA-F and the SAF-SP is within a predetermined threshold, and the SAF-SP is at a maximum limit.

18. The method of claim 13 , wherein the determining the one or more VAVs as a suspect VAV comprises employing a machine learning model.

19. A building management system comprising:

one or more variable air volume units (VAVs) configured to provide air flow into a building;

an air handling unit (AHU) configured to provide the air flow to the one or more VAVs; and

one or more memory devices configured to store instructions thereon that, when executed by one or more processors, cause the one or more processors to:

receive VAV data from the one or more of VAVs or AHU data from the AHU;

determine that one or more VAVs as a suspect VAV in response to one or more conditions being satisfied by at least one of (i) the VAV data or (ii) the VAV data and the AHU data;

classify the suspect VAV as a first type of suspect VAV in response to a first set of the one or more conditions being satisfied;

classify the suspect VAV as a second type of suspect VAV in response to a second set of the one or more conditions being satisfied

in response to the suspect VAV being classified as the first type of suspect VAV, remove a portion of the VAV data associated with the suspect VAV from a calculation of a duet static pressure setpoint for the AHU and operate building equipment to affect an airflow provided to one or more building zones using an adjusted value of the duct static pressure setpoint after removing the portion of the VAV data associated with the suspect VAV from the calculation of the duct static pressure setpoint; and

in response to the suspect VAV being classified as the second type of suspect VAV, increase or decrease the value of the DPT-SP and operate the building equipment to affect the airflow provided to the one or more building zones using the increased or decreased value of the DPT-SP.

20. The building management system of claim 19 , where the one or more processors employs a machine learning model for tagging the suspect VAV.