Data-driven demand charge management solution

1. A computer-implemented method for demand charge management, comprising:

receiving an active power demand for a facility, a current load demand charge threshold (DCT) profile for the facility, and a plurality of previously observed load DCT profiles;

generating a data set of DCT values based on the current load DCT profile for the facility and the plurality of previously observed load DCT profiles;

forecasting a forecast DCT value for a next month for the facility using the data set of DCT values by adding a DCT value for the next month from a previous year to a summation of DCT values in the data set of DCT values and then dividing by a number of DCT values in the data set of DCT values plus 1; and

preventing actual power used from a utility from exceeding the forecast DCT value by discharging a battery storage system into a behind the meter power infrastructure for the facility.

2. The computer-implemented method of claim 1 , further comprising charging the battery storage system when the active power demand is less than the forecast DCT.

3. The computer-implemented method of claim 2 , wherein said charging step comprises confirming a current state of charge is less than a maximum state of charge.

4. The computer-implemented method of claim 1 , wherein said discharging step occurs when a current state of charge is greater than or equal to a minimum state of charge.

5. The computer-implemented method of claim 1 , wherein the plurality of previously observed load DCT profiles include profiles from the facility and other facilities.

6. The computer-implemented method of claim 1 , wherein said forecasting step comprises computing the Euclidean distance between the current DCT profile and one or more of the plurality of previously observed load DCT profiles in the data set of DCT values.

7. The computer-implemented method of claim 6 , wherein said forecasting step comprises generating a similarity group of DCT profiles including the current DCT profile and the one or more of the plurality of previously observed load DCT profiles with computed Euclidean distances less than a threshold value.

8. The computer-implemented method of claim 7 , wherein said forecasting step comprises normalizing DCT values of DCT profiles in the similarity group of DCT profiles.

9. The computer-implemented method of claim 8 , wherein said forecasting step uses an average of the normalized DCT values of DCT profiles in the similarity group of DCT profiles.

10. A computer program product for demand charge management, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a method comprising:

receiving an active power demand for a facility, a current load demand charge threshold (DCT) profile for the facility, and a plurality of previously observed load DCT profiles;

generating a data set of DCT values based on the current load DCT profile for the facility and the plurality of previously observed load DCT profiles;

forecasting a forecast DCT value for a next month for the facility using the data set of DCT values by adding a DCT value for the next month from a previous year to a summation of DCT values in the data set of DCT values and then dividing by a number of DCT values in the data set of DCT values plus 1; and

preventing actual power used from a utility from exceeding the forecast DCT value by discharging a battery storage system into a behind the meter power infrastructure for the facility.

11. A demand charge management system, comprising:

a processor configured to:

receive an active power demand for a facility, a current load demand charge threshold (DCT) profile for the facility, and a plurality of previously observed load DCT profiles;

generate a data set of DCT values based on the current load DCT profile for the facility and the plurality of previously observed load DCT profiles;

forecast a forecast DCT value for a next month for the facility using the data set of DCT values by adding a DCT value for the next month from a previous year to a summation of DCT values in the data set of DCT values and then dividing by a number of DCT values in the data set of DCT values plus 1; and

prevent actual power used from a utility from exceeding the forecast DCT by discharging a battery storage system into a behind the meter power infrastructure for the facility.

12. The system of claim 11 , wherein the processor is further configured to charge the battery storage system when the active power demand is less than the forecast DCT.

13. The system of claim 12 , wherein the processor is further configured to charge the battery storage system after confirming a current state of charge is less than a maximum state of charge.

14. The system of claim 11 , wherein the processor is further configured to discharge the battery storage system occurs when a current state of charge is greater than or equal to a minimum state of charge.

15. The system of claim 11 , wherein the plurality of previously observed load DCT profiles include profiles from the facility and other.

16. The system of claim 11 , wherein the processor forecasts the next month DCT value by computing the Euclidean distance between the current DCT profile and one or more of the plurality of previously observed load DCT profiles in the data set of DCT values.

17. The system of claim 16 , wherein the processor forecasts the next month DCT value by generating a similarity group of DCT profiles including the current DCT profile and the one or more of the plurality of previously observed load DCT profiles with computed Euclidean distances less than a threshold value.

18. The system of claim 17 , wherein the processor forecasts the next month DCT value by normalizing DCT values of DCT profiles in the similarity group of DCT profiles.

19. The system of claim 18 , wherein the processor forecasts the next month DCT value by using an average of the normalized DCT values of DCT profiles in the similarity group of DCT profiles.

20. The system of claim 11 , wherein the plurality of previously observed load DCT profiles includes profiles for the facility that span one or more time periods.