Cognitive energy assessment by a non-intrusive sensor in a thermal energy fluid transfer system

1. A method for assessing energy in a thermal energy fluid transfer system in a cloud computing environment by a processor, comprising:

deploying a high frequency sampling operation in the thermal energy fluid transfer system associated with a heating service, a cooling service, or a combination thereof by identifying and accessing a data object within a data historian warehouse of existing resource infrastructure associated with the thermal energy fluid transfer system, and initiating the high frequency sampling operation using the data object, wherein the high frequency sampling operation samples data at a predetermined frequency for a predetermined time period, the predetermined time period comprising a time period during which a threshold amount of data is collected to use as training data as input to a machine learning operation modeling the thermal energy fluid transfer system, and wherein the data collected by the high frequency sampling operation is obtained by identifying a temperature of fluid transferred through the thermal energy fluid transfer system at the predetermined frequency;

executing machine learning logic performing the machine learning operation to use the training data to learn behavior of the thermal energy fluid transfer and generate tuning threshold levels for the modeling of the thermal energy fluid transfer system according to the learned behavior; and

performing energy usage assessment by applying a disambiguation operation to a single temperature signal, representing the temperature of the fluid and collected over a second predetermined time period by only a single non-intrusive Internet of Things (IoT) sensor coupled externally to a return piping section at a selected position in the thermal energy fluid transfer system, for one or more types of events to identify one or more energy usage waste events according to the learned behavior, wherein the disambiguation operation includes performing real-time statistical detection operations to identify a system-wide usage pattern of the thermal energy fluid transfer system using collected time-series data from only a single data point by the single non-intrusive IoT sensor, wherein the selected position includes coupling the single non-intrusive IoT sensor externally to the return piping at a selectively measured distance from a defined reference point of the thermal energy fluid transfer system, and wherein the single non-intrusive IoT sensor is externally coupled to the return piping comprising a loop of pipe network located subsequent to an energy load serviced by the thermal energy fluid transfer system to continuously sample temperature data to detect load and no load actions of the one or more energy usage waste events.

2. The method of claim 1 , wherein the disambiguation operation further includes:

dissecting the temperature signal into the one or more types of events; and

analyzing the temperature signal to obtain a signature for the one or more type of events.

3. The method of claim 1 , further including generating one or more energy usage tuning thresholds according to the collected data.

4. The method of claim 1 , further including comparing the temperature signal of one or more current energy usage events to tuning threshold levels to identify the one or more energy usage waste events.

5. The method of claim 1 , wherein the machine learning operation uses feedback information of the data collected from the single non-intrusive IoT sensor to provide an energy usage profile of the thermal energy fluid transfer system.

6. The method of claim 1 , further including:

determining a thermal energy fluid transfer system start time and a thermal energy fluid transfer system stop time for a schedule of operations to identify the one or more energy usage waste events;

estimating energy required events and non-energy required events during the schedule of operations; and

determining an unexpected change point during the schedule of operations, wherein a change point is a transition from the energy required events to the non-energy required events.

7. The method of claim 1 , further including:

providing an alert to indicate identification of the one or more energy usage waste events;

providing the energy usage assessment to a user via an interactive graphical user interface (GUI) for the thermal energy fluid transfer system.

8. A system for assessing energy in a thermal energy fluid transfer system in a cloud computing environment, comprising:

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

deploy a high frequency sampling operation in the thermal energy fluid transfer system associated with a heating service, a cooling service, or a combination thereof by identifying and accessing a data object within a data historian warehouse of existing resource infrastructure associated with the thermal energy fluid transfer system, and initiating the high frequency sampling operation using the data object, wherein the high frequency sampling operation samples data at a predetermined frequency for a predetermined time period, the predetermined time period comprising a time period during which a threshold amount of data is collected to use as training data as input to a machine learning operation modeling the thermal energy fluid transfer system, and wherein the data collected by the high frequency sampling operation is obtained by identifying a temperature of fluid transferred through the thermal energy fluid transfer system at the predetermined frequency;

execute machine learning logic performing the machine learning operation to use the training data to learn behavior of the thermal energy fluid transfer and generate tuning threshold levels for the modeling of the thermal energy fluid transfer system according to the learned behavior; and

perform energy usage assessment by applying a disambiguation operation to a single temperature signal, representing the temperature of the fluid and collected over a second predetermined time period by only a single non-intrusive Internet of Things (IoT) sensor coupled externally to a return piping section at a selected position in the thermal energy fluid transfer system, for one or more types of events to identify one or more energy usage waste events according to the learned behavior, wherein the disambiguation operation includes performing real-time statistical detection operations to identify a system-wide usage pattern of the thermal energy fluid transfer system using collected time-series data from only a single data point by the single non-intrusive IoT sensor, wherein the selected position includes coupling the single non-intrusive IoT sensor externally to the return piping at a selectively measured distance from a defined reference point of the thermal energy fluid transfer system, and wherein the single non-intrusive IoT sensor is externally coupled to the return piping comprising a loop of pipe network located subsequent to an energy load serviced by the thermal energy fluid transfer system to continuously sample temperature data to detect load and no load actions of the one or more energy usage waste events.

9. The system of claim 8 , wherein the executable instructions further:

use feedback information of the collected data from the single non-intrusive IoT sensor to provide an energy usage profile of the thermal energy fluid transfer system by the machine learning operation; and

generate one or more energy usage tuning thresholds according to the collected data and feedback information.

10. The system of claim 8 , wherein the disambiguation operation further includes executable instructions that:

dissect the temperature signal into the one or more types of events; and

analyze the temperature signal to obtain a signature for the one or more type of events.

11. The system of claim 8 , wherein the executable instructions further compare the temperature signal of one or more current energy usage events to the tuning threshold levels to identify the one or more energy usage waste events.

12. The system of claim 8 , wherein the executable instructions further:

determine a thermal energy fluid transfer system start time and a thermal energy fluid transfer system stop time for a schedule of operations to identify the one or more energy usage waste events;

estimate energy required events and non-energy required events during the schedule of operations; and

determine an unexpected change point during the schedule of operations, wherein a change point is a transition from the energy required events to the non-energy required events.

13. The system of claim 8 , wherein the executable instructions further:

provide an alert to indicate identification of the one or more energy usage waste events; and

provide the energy usage assessment to a user via an interactive graphical user interface (GUI) for the thermal energy fluid transfer system.

14. A computer program product for assessing energy in a thermal energy fluid transfer system in a cloud 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 deploys a high frequency sampling operation in the thermal energy fluid transfer system associated with a heating service, a cooling service, or a combination thereof by identifying and accessing a data object within a data historian warehouse of existing resource infrastructure associated with the thermal energy fluid transfer system, and initiating the high frequency sampling operation using the data object, wherein the high frequency sampling operation samples data at a predetermined frequency for a predetermined time period, the predetermined time period comprising a time period during which a threshold amount of data is collected to use as training data as input to a machine learning operation modeling the thermal energy fluid transfer system, and wherein the data collected by the high frequency sampling operation is obtained by identifying a temperature of fluid transferred through the thermal energy fluid transfer system at the predetermined frequency;

an executable portion that executes machine learning logic performing the machine learning operation to use the training data to learn behavior of the thermal energy fluid transfer and generate tuning threshold levels for the modeling of the thermal energy fluid transfer system according to the learned behavior;

an executable portion that performs energy usage assessment by applying a disambiguation operation to a single temperature signal, representing the temperature of the fluid and collected over a second predetermined time period by only a single non-intrusive Internet of Things (IoT) sensor coupled externally to a return piping section at a selected position in the thermal energy fluid transfer system, for one or more types of events to identify one or more energy usage waste events according to the learned behavior, wherein the disambiguation operation includes performing real-time statistical detection operations to identify a system-wide usage pattern of the thermal energy fluid transfer system using collected time-series data from only a single data point by the single non-intrusive IoT sensor, wherein the selected position includes coupling the single non-intrusive IoT sensor externally to the return piping at a selectively measured distance from a defined reference point of the thermal energy fluid transfer system, and wherein the single non-intrusive IoT sensor is externally coupled to the return piping comprising a loop of pipe network located subsequent to an energy load serviced by the thermal energy fluid transfer system to continuously sample temperature data to detect load and no load actions of the one or more energy usage waste events.

15. The computer program product of claim 14 , further including an executable portion that:

uses feedback information of the collected data from the single non-intrusive IoT sensor to provide an energy usage profile of the thermal energy fluid transfer system by the machine learning operation; and

generates one or more energy usage tuning thresholds according to the collected data and feedback information.

16. The computer program product of claim 14 , wherein the disambiguation operation further includes an executable portion that:

dissects the temperature signal into the one or more types of events; and

analyzes the temperature signal to obtain a signature for the one or more type of events.

17. The computer program product of claim 14 , further including an executable portion that compares the temperature signal of one or more current energy usage events to the tuning threshold levels to identify the one or more energy usage waste events.

18. The computer program product of claim 14 , further including an executable portion that:

determines a thermal energy fluid transfer system start time and a thermal energy fluid transfer system stop time for a schedule of operations to identify the one or more energy usage waste events;

estimates energy required events and non-energy required events during the schedule of operations; and

determines an unexpected change point during the schedule of operations, wherein a change point is a transition from the energy required events to the non-energy required events.

19. The computer program product of claim 14 , further including an executable portion that provides an alert to indicate identification of the one or more energy usage waste events.

20. The computer program product of claim 14 , further including an executable portion that provides the energy usage assessment to a user via an interactive graphical user interface (GUI) for the thermal energy fluid transfer system.