Method and system for anomaly detection, missing data imputation and consumption prediction in energy data

1. A method for outlier detection, anomalous behavior detection, missing data imputation and prediction of consumption in energy data for one or more energy sensors by using a unified model, said method comprising processor implemented steps of:

collecting a time series data to be used as training data, wherein the time series data is generated and modeled according to a preferred interval;

training the unified model using the collected time series data to enable computation of a plurality of parameters, wherein the plurality of parameters are computed by implementing statistical machine learning techniques on the collected time series data, wherein the statistical machine learning techniques uses Inter Quartile Ranges (IQR) to calculate normal consumption ranges and thresholds for each data points, wherein the thresholds are identified at different levels of hierarchy and combined with occupancy to obtain patterns for usage per occupant for providing precise levels of consumption, and wherein the unified model is periodically re-trained to stabilize the patterns, making the unified model self-learning and self-optimizing; and

implementing using the trained unified model the plurality of parameters on a new data of energy consumption, wherein the plurality of parameters are used to perform the outlier detection, anomaly detection, missing data imputation and prediction of consumption in energy data.

2. The method according to claim 1 , further comprising calibrating the one or more energy sensors based on the implementation of the trained unified model on the new data of energy consumption.

3. The method according to claim 1 , further comprising reporting the outlier detection, anomalous behavior detection, missing data imputation and prediction of consumption in energy data for the one or more energy sensors.

4. The method according to claim 1 , wherein the plurality of parameters are saved in a database such that the database is updated periodically to implement the data.

5. The method according to claim 1 , wherein the collected time series data used as training data is un-cleaned data.

6. A system ( 102 ) for outlier detection, anomalous behavior detection, missing data imputation and prediction of consumption in energy data for one or more energy sensors by using a unified model, the system comprising:

a processor ( 202 ),

a memory ( 204 ), operatively coupled with said processor, wherein the processor is configured to:

collect a time series data to be used as training data, wherein the time series data is generated and modeled according to a preferred interval;

train the unified model using the collected time series data to enable computation of a plurality of parameters wherein the plurality of parameters are computed by implementing statistical machine learning techniques on the collected time series data, wherein the statistical machine learning techniques uses Inter Quartile Ranges (IQR) to calculate normal consumption ranges and thresholds for each data points, wherein the thresholds are identified at different levels of hierarchy and combined with occupancy to obtain patterns for usage per occupant for providing precise levels of consumption, and wherein the unified model is periodically re-trained to stabilize the patterns, making the unified model self-learning and self-optimizing; and

implement, using the trained unified model, the plurality of parameters on a new data of energy consumption wherein the plurality of parameters are used to perform the outlier detection, anomaly detection, missing data imputation and prediction of consumption in energy data.

7. The system according to claim 6 , further comprising calibrating the one or more energy sensors based on the implementation of the trained unified model on the new data of energy consumption.

8. The system according to claim 6 , further comprising reporting the outlier detection, anomalous behavior detection, missing data imputation and prediction of consumption in energy data for the one or more energy sensors.

9. The system according to claim 6 , wherein the plurality of parameters are saved in a database such that the database is updated periodically to implement the data.

10. One or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:

collecting a time series data to be used as training data using a data collection module ( 210 ), wherein the time series data is generated and modeled according to a preferred interval;

training a unified model using the collected time series data to enable computation of a plurality of parameters, wherein the plurality of parameters are computed by implementing statistical machine learning techniques on the collected time series data, wherein the statistical machine learning techniques uses Inter Quartile Ranges (IQR) to calculate normal consumption ranges and thresholds for each data points, wherein the thresholds are identified at different levels of hierarchy and combined with occupancy to obtain patterns for usage per occupant for providing precise levels of consumption, and wherein the unified model is periodically re-trained to stabilize the patterns, making the unified model self-learning and self-optimizing; and

implementing using the trained unified model the plurality of parameters on a new data of energy consumption, wherein the plurality of parameters are used to perform the outlier detection, anomaly detection, missing data imputation and prediction of consumption in energy data.

11. The one or more non-transitory machine readable information storage mediums of claim 10 , wherein the one or more instructions which when executed by the one or more hardware processors further cause calibrating the one or more energy sensors based on the implementation of the trained unified model on the new data of energy consumption using a calibration module ( 216 ).

12. The one or more non-transitory machine readable information storage mediums of claim 10 , wherein the one or more instructions which when executed by the one or more hardware processors further cause reporting the outlier detection, anomalous behavior detection, missing data imputation and prediction of consumption in energy data for the one or more energy sensors.

13. The one or more non-transitory machine readable information storage mediums of claim 10 , wherein the plurality of parameters are saved in a database such that the database is updated periodically to implement the data.

14. The one or more non-transitory machine readable information storage mediums of claim 10 , wherein the collected time series data used as training data is un-cleaned data.