Systems and methods for adaptively tuning thresholds for fault detection in buildings

1. A building system, comprising one or more memory devices configured to store instructions that, when executed on one or more processors, cause the one or more processors to:

determine an average of a minimum half of sorted energy consumption values for a first time period;

determine an average of a maximum half of sorted energy consumption values for a second time period;

determine a ratio of the average of the minimum half of sorted energy consumption values for the first time period to the average of the maximum half of sorted energy consumption values for the second time period;

compare the calculated ratio to a threshold value; and

control a second system responsive to the calculated ratio exceeding the threshold value, wherein the controlled second system controls one or more building subsystems, one or more building management system devices, or one or more parameters for controlling components of an HVAC system.

2. The building system of claim 1 , wherein the first time period is an occupancy time period and the second time period is a non-occupancy time period.

3. The building system of claim 1 , wherein the instructions cause the one or more processors to:

determine a mean of a set of calculated ratios for an interval;

determine a median absolute deviation of the calculated ratios for the interval; and

set the threshold value based on the mean and the median absolute deviation.

4. The building system of claim 3 , wherein the threshold value is further based on a scaling factor to scale the median absolute deviation based on the mean of the set of calculated ratios.

5. The building system of claim 1 , wherein the instructions cause the one or more processors to:

collect data samples from one or more building management system devices and generate a data timeseries comprising a plurality of the data samples;

perform one or more analytics using the data timeseries and generate a results timeseries comprising a plurality of result samples indicating results of the analytics, wherein the building system further comprises a database configured to store the data timeseries and the results timeseries; and

retrieve the data timeseries and the results timeseries from the database in response to a request for timeseries data associated with one or more variables.

6. The building system of claim 5 , wherein the instructions further cause the one or more processors to generate the results timeseries by removing an effect of weather from the data timeseries.

7. The building system of claim 5 , wherein the data timeseries is a resource consumption timeseries and the data samples of the data timeseries comprise at least one of electric consumption values, water consumption values, or natural gas consumption values; and

wherein the instructions further cause the one or more processors to use the data timeseries to calculate an energy usage metric for a building associated with the data timeseries, the energy usage metric comprising at least one of energy usage intensity (EUI) or energy density.

8. The building system of claim 5 , wherein the instructions further cause the one or more processors to fill one or more gaps in data of the data timeseries using values from a third time period different from the first time period and the second time period.

9. The building system of claim 8 , wherein the third time period and the first time period are associated with a common characteristic.

10. A system comprising:

a building management system in communication with one or more building management system devices of a building network,

wherein the building management system is configured to:

determine an average of energy consumption values for a first time period;

determine an average of energy consumption values for a second time period;

determine a ratio of the average energy consumption values for the first time period to the average of energy consumption values for the second time period;

determine a mean of a set of calculated ratios for an interval;

determine a median absolute deviation of the calculated ratios for the interval;

determine an adaptively tunable threshold value based on the mean and the median absolute deviation;

compare the calculated ratio to the adaptively tunable threshold value; and

control a second system responsive to the calculated ratio exceeding the adaptively tunable threshold value, wherein the controlled second system controls one or more building subsystems, one or more building management system devices, or one or more parameters for controlling components of an HVAC system.

11. The system of claim 10 , wherein the first time period is a weekend time period and the second time period is a weekday time period.

12. The system of claim 10 , wherein the adaptively tunable threshold value is further based on a scaling factor to scale the median absolute deviation based on the mean of the set of calculated ratios.

13. The system of claim 10 , wherein the building management system is further configured to:

collect data samples from the one or more building management system devices and generate a data timeseries comprising a plurality of the data samples;

perform one or more analytics using the data timeseries and generate a results timeseries comprising a plurality of result samples indicating results of the analytics, wherein the system further comprises a database configured to store the data timeseries and the results timeseries; and

retrieve the data timeseries and the results timeseries from the database in response to a request for timeseries data associated with one or more variables.

14. The system of claim 13 , wherein the building management system is further configured to generate the results timeseries by removing an effect of weather from the data timeseries.

15. The system of claim 13 , wherein the data timeseries is a resource consumption timeseries and the data samples of the data timeseries comprise at least one of electric consumption values, water consumption values, or natural gas consumption values; and

wherein the building management system is further configured to use the data timeseries to calculate an energy usage metric for a building associated with the data timeseries, the energy usage metric comprising at least one of energy usage intensity (EUI) or energy density.

16. The system of claim 13 , wherein the building management system is further configured to fill one or more gaps in data of the data timeseries using values from a third time period different from the first time period and the second time period.

17. A method conducted by a processing circuit of a building management system, comprising:

determining, by the processing circuit, an average of a minimum half of sorted energy consumption values for a first time period;

determining, by the processing circuit, an average of a maximum half of sorted energy consumption values for a second time period;

determining, by the processing circuit, a ratio of the average of the minimum half of sorted energy consumption values for the first time period to the average of the maximum half of sorted energy consumption values for the second time period;

comparing, by the processing circuit, the determined ratio to a threshold value; and

controlling, by the processing circuit, a system responsive to the determined ratio exceeding the adaptively tunable threshold value, wherein the controlled second system controls one or more building subsystems, one or more building management system devices, or one or more parameters for controlling components of an HVAC system.

18. The method of claim 17 , further comprising:

collecting, by the processing circuit, data samples from building management system devices of the building management system and generating a data timeseries comprising a plurality of the data samples;

performing, by the processing circuit, one or more analytics using the data timeseries;

generating, by the processing circuit, a results timeseries comprising a plurality of result samples indicating results of the analytics;

storing, by the processing circuit, the data timeseries and the results timeseries;

retrieving, by the processing circuit, the data timeseries and the results timeseries from a database in response to a request for timeseries data associated with the one or more variables,

wherein determining the average of a minimum half of sorted energy consumption values for the first time period and the average of a maximum half of sorted energy consumption values for the second time period is performed using samples of the data timeseries.