Localizing energy consumption anomalies in buildings

1. A method for localizing abnormal energy consumption at a facility by a processor, comprising:

monitoring, by the processor, data collected by one or more Internet of Things (IoT) devices, wherein the data is representative of a combination of one or more energy consumption measurements, weather data, and one or more characteristics of one or more facilities;

training, by the processor, one or more machine learning models using the monitored data, wherein the machine learning model is used to tune combinations of features or parameters of the one or more IoT devices to enable the collection of the monitored data;

generating, by the processor, one or more residuals for both one or more predictors and a prediction according to an analyzation of the monitored data by the trained one or more machine learning models;

localizing, by the processor based on the analyzation, an energy consumption anomaly to a particular meter or sub-meter within a specific zone of the one or more facilities according to those of the one or more residuals associated with one or more predictors having an actual energy measurement deviating from a predicted actual energy measurement, wherein the predictors are sub-meters monitoring a plurality of zones of each of the one or more facilities and the prediction is a meter monitoring each of the sub-meters of the one or more facilities, and wherein localizing the energy consumption anomaly is performed by correlating the one or more residuals of the sub-meters to the one or more residuals of the meter, including compensating for missing and outlying values of ones of the sub-meters or meter, to determine the particular meter or sub-meter within the specific zone the energy consumption anomaly is detected within; and

automatically alerting, by the processor, the one or more IoT devices of the localized energy consumption anomaly via a display of the one or more IoT devices.

2. The method of claim 1 , further including detecting the energy consumption anomaly for the one or more facilities according to previously determined residuals.

3. The method of claim 1 , wherein monitoring the energy consumption further includes using a data cleaning operation, a forecasting operation to predict energy consumption so as to generate the one or more residuals, a change point detection operation, or a combination thereof.

4. The method of claim 1 , further including assigning a confidence level to each of the one or more residuals.

5. The method of claim 1 , further including

determining a contribution level of each of the one or more predictors to a total energy consumption level of the one or more facilities according to the correlating.

6. A system for localizing abnormal energy consumption at a facility in a computing environment, comprising:

one or more computers with executable instructions that when executed cause the system to:

monitors, by a processor associated with the one or more computers and executing the executable instructions, data collected by one or more Internet of Things (IoT) devices, wherein the data is representative of a combination of one or more energy consumption measurements, weather data, and one or more characteristics of one or more facilities;

trains, by the processor, one or more machine learning models using the monitored data, wherein the machine learning model is used to tune combinations of features or parameters of the one or more IoT devices to enable the collection of the monitored data;

generate, by the processor, one or more residuals for both one or more predictors and a prediction according to an analyzation of the monitored data by the trained one or more machine learning models;

localize, by the processor based on the analyzation, an energy consumption anomaly to a particular meter or sub-meter within a specific zone of the one or more facilities according to those of the one or more residuals associated with one or more predictors having an actual energy measurement deviating from a predicted actual energy measurement, wherein the predictors are sub-meters monitoring a plurality of zones of each of the one or more facilities and the prediction is a meter monitoring each of the sub-meters of the one or more facilities, and wherein localizing the energy consumption anomaly is performed by correlating the one or more residuals of the sub-meters to the one or more residuals of the meter, including compensating for missing and outlying values of ones of the sub-meters or meter, to determine the particular meter or sub-meter within the specific zone the energy consumption anomaly is detected within; and

automatically alert, by the processor, the one or more IoT devices of the localized energy consumption anomaly via a display of the one or more IoT devices.

7. The system of claim 6 , wherein the executable instructions further detect the energy consumption anomaly for the one or more facilities according to previously determined residuals.

8. The system of claim 6 , wherein monitoring the energy consumption further includes using a data cleaning operation, a forecasting operation to predict energy consumption so as to generate the one or more residuals, a change point detection operation, or a combination thereof.

9. The system of claim 6 , wherein the executable instructions further assign a confidence level to each of the one or more residuals.

10. The system of claim 6 , wherein the executable instructions further

determine a contribution level of each of the one or more predictors to a total energy consumption level of the one or more facilities according to the correlating.

11. A computer program product for detection of energy consumption in a building associated with a computing environment by a processor, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

an executable portion that monitors, by the processor, data collected by one or more Internet of Things (IoT) devices, wherein the data is representative of a combination of one or more energy consumption measurements, weather data, and one or more characteristics of one or more facilities;

an executable portion that trains, by the processor, one or more machine learning models using the monitored data, wherein the machine learning model is used to tune combinations of features or parameters of the one or more IoT devices to enable the collection of the monitored data;

an executable portion that generates, by the processor, one or more residuals for both one or more predictors and a prediction according to an analyzation of the monitored data by the trained one or more machine learning models;

an executable portion that localizes, by the processor based on the analyzation, an energy consumption anomaly to a particular meter or sub-meter within a specific zone of the one or more facilities according to those of the one or more residuals associated with one or more predictors having an actual energy measurement deviating from a predicted actual energy measurement, wherein the predictors are sub-meters monitoring a plurality of zones of each of the one or more facilities and the prediction is a meter monitoring each of the sub-meters of the one or more facilities, and wherein localizing the energy consumption anomaly is performed by correlating the one or more residuals of the sub-meters to the one or more residuals of the meter, including compensating for missing and outlying values of ones of the sub-meters or meter, to determine the particular meter or sub-meter within the specific zone the energy consumption anomaly is detected within; and

an executable portion that automatically alerts, by the processor, the one or more IoT devices of the localized energy consumption anomaly via a display of the one or more IoT devices.

12. The computer program product of claim 11 , further including an executable portion that detects the energy consumption anomaly for the one or more facilities according to previously determined residuals.

13. The computer program product of claim 11 , wherein monitoring the energy consumption further includes using a data cleaning operation, a forecasting operation to predict energy consumption so as to generate the one or more residuals, a change point detection operation, or a combination thereof.

14. The computer program product of claim 11 , further including an executable portion that assigns a confidence level to each of the one or more residuals.

15. The computer program product of claim 11 , further including an executable portion that

determines a contribution level of each of the one or more predictors to a total energy consumption level of the one or more facilities according to the correlating.