Systems and methods for intervention control in a building management system

1. A method of predicting a time of effect of an intervention of a point of a Building Management System (BMS), the method comprising:

receiving, from a user interface, an input request for an intervention of a point of the BMS, the intervention changing a setpoint of the point from a first value to a second value;

predicting a time period in which the intervention will cause a variable of the BMS to be within a predetermined range;

presenting feedback via the user interface before implementing the intervention of the point, the feedback comprising the time period in which the intervention will cause the variable to be within the predetermined range; and

using an electronic processor, implementing the intervention based at least in part on an input from the user or an automated response to the feedback by:

changing, using the electronic processor, the setpoint of the point from the first value to the second value;

responsive to the change, changing, using the electronic processor, operation of building equipment of the BMS according to the changed setpoint;

reverting, using the electronic processor, the setpoint to the first value within the predicted time period; and

responsive to the reversion, changing, using the electronic processor, the operation of the building equipment according to the reverted setpoint.

2. The method of claim 1 ,

wherein predicting the time period in which the intervention will cause the variable of the BMS to be within the predetermined range comprises determining a first time at which the intervention will cause a third value of the variable to reach the predetermined range.

3. The method of claim 2 , further comprising:

wherein predicting the time period in which the intervention will cause the variable of the BMS to be within the predetermined range comprises determining a second time at which the intervention will cause a fourth value of the variable to leave the predetermined range.

4. The method of claim 1 , wherein predicting the time period in which the intervention will cause the variable of the BMS to be within the predetermined range comprises:

obtaining live BMS data associated with operation of the BMS;

providing the live BMS data to one or more machine learning models;

obtaining a confidence score for a value for the variable for a first time within the time period by executing the one or more machine learning models using the live BMS data, the confidence score indicating a degree of confidence that the value is correct for the first time; and

predicting the time period based on the confidence score.

5. The method of claim 4 , further comprising:

generating a training dataset using historical data associated with the BMS, the training dataset comprising timeseries with an output indicating a second time that a second intervention will affect the variable; and

training the one or more machine learning models using the training dataset.

6. The method of claim 4 , wherein the one or more machine learning models comprise at least one of:

a first machine learning model for predicting control system data if there is not an intervention; or

a second machine learning model for predicting the control system data if there is an intervention,

wherein predicting the time period in which the intervention will cause the variable to be within the predetermined range is based on an output from the second machine learning model.

7. The method of claim 6 , wherein the one or more machine learning models further comprise at least one of:

a third machine learning model for predicting an effect that external loading disturbances will have on the BMS; or

a fourth machine learning model for predicting an effect that internal loading disturbances will have on the BMS,

wherein the at least one of the third machine learning model or the fourth machine learning model provide outputs to the at least one of the first machine learning model or the second machine learning model.

8. The method of claim 1 , wherein the variable is a comfortability of the BMS or an energy or cost savings of the BMS.

9. The method of claim 1 , further comprising:

concurrently presenting, on the user interface, first values of the variable over time if the user does not input the intervention and second values of the variable over time if the user does input the intervention.

10. The method of claim 1 , wherein the variable is an energy consumption, a zone air temperature, a zone humidity, or a discharge air temperature.

11. The method of claim 1 , wherein the point is a supply vent state, a supply fan state, a discharge air temperature setpoint, a zone air temperature setpoint, a ductwork damper state, or a cooling coil valve position.

12. A system comprising one or more memory devices storing instructions thereon that, when executed by one or more processors, cause the one or more processors to:

receive a first input related to an intervention of a point, the intervention changing a setpoint of the point from a first value to a second value;

predict a time period in which the intervention will cause a variable of the system to be within a predetermined range; and

implement the intervention the intervention based at least in part on the time period by:

changing the setpoint of the point from the first value to the second value;

responsive to the change, changing operation of building equipment of the BMS according to the changed setpoint;

reverting the setpoint to the first value within the predicted time period; and

responsive to the reversion, changing the operation of the building equipment according to the reverted setpoint.

13. The system of claim 12 , wherein the instructions cause the one or more processors to predict the time period in which the intervention will cause the variable of the system to be within the predetermined range by determining a first time at which the intervention will cause a third value of the variable to reach the predetermined range.

14. The system of claim 13 , wherein the instructions cause the one or more processors to predict the time period in which the intervention will cause the variable of the system to be within the predetermined range by determining a second time at which the intervention will cause a fourth value of the variable to leave the predetermined range.

15. The system of claim 12 , wherein the instructions cause the one or more processors to predict the time period in which the intervention will cause the variable of the system to be within the predetermined range by:

obtaining live system data associated with operation of the system;

providing the live system data to one or more machine learning models;

obtaining a confidence score for a value for the variable for a first time within the time period by executing the one or more machine learning models using the live system data, the confidence score indicating a degree of confidence that the value is correct for the first time; and

predicting the time period based on the confidence score.

16. The method of claim 15 , wherein the instructions cause the one or more processors to further:

generate a training dataset using historical data associated with the BMS, the training dataset comprising timeseries with an output indicating a second time at which a second intervention will affect the variable; and

train the one or more machine learning models using the training dataset.

17. A method for predicting a time of effect of an intervention of a point of a Building Management System (BMS), the method comprising:

predicting a time period in which an intervention of a point will cause a value of a variable to be to be within a predetermined range, the intervention changing a setpoint of the point from a first value to a second value;

presenting feedback via a user interface before implementing the intervention of the point, the feedback comprising the time period in which the intervention will cause the variable to be within the predetermined range; and

using an electronic processor, implementing the intervention or a modification of the intervention based at least in part on an input from the user or an automated response to the feedback by:

changing, by the electronic processor, the setpoint of the point from the first value to the second value;

responsive to the change, changing, by the electronic processor, operation of building equipment of the BMS according to the changed setpoint;

reverting, using the electronic processor, the setpoint to the first value within the predicted time period; and

responsive to the reversion, changing, using the electronic processor, the operation of the building equipment according to the reverted setpoint.

18. The method of claim 17 , wherein predicting the time period in which the intervention will cause the variable of the BMS to be within the predetermined range comprises:

obtaining live BMS data associated with operation of the BMS;

providing the live BMS data to one or more machine learning models;

obtaining a confidence score for a value for the variable for a first time within the time period by executing the one or more machine learning models using the live BMS data, the confidence score indicating a degree of confidence that the value is correct for the first time; and

predicting the time period based on the confidence score.

19. The method of claim 18 , wherein the one or more machine learning models comprise at least one selected of:

a first machine learning model for predicting control system data if there is not an intervention; or

a second machine learning model for predicting control system data if there is an intervention,

wherein predicting a time period in which the intervention will cause the variable of the BMS to be within the predetermined range is based on an output from the second machine learning model.

20. The method of claim 1 , wherein the time period begins at a first time and ends at a second time,

wherein changing the setpoint of the point from the first value to the second value comprises changing, using the electronic processor, the setpoint prior to the first time; and

wherein reverting the setpoint to the first value within the predicted time period comprises reverting, using the electronic processor, the setpoint to the first value prior to the second time.