Predictive smart grid re-structuring based on expected load and power generation

• US 9,389,630 B2
• Filed: 02/21/2014
• Issued: 07/12/2016
• Est. Priority Date: 02/21/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kalagnanam, Jayant R, Gaucher, Brian P., Mukherjee, Maharaj, Treinish, Lloyd A.
• Primary Examiner(s): Mitchell, Jason
• Application Number:
  US14/186,361
• Publication Number:
  US 20150241892A1
• Time in Patent Office: 872 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G05B 13/026 using a predictor
• CPC Class Codes: G05F 1/66 Regulating electric power
• CPC Class Codes: G05F 1/67 to the maximum power available from a generator, e.g. from solar cell
• CPC Class Codes: G06Q 50/06 Energy or water supply
• CPC Class Codes: Y02A 30/00 Adapting or protecting infrastructure or their operation

Abstract

A method for structuring an electric utility grid having traditional and renewable sources of electric power uses a weather forecasting system to determine, for a time interval, whether there are paths between all generating units and all consuming units satisfying certain constraints. The method includes computing a probable output for each of the generating units, both traditional and renewable, and computing a probable load for each of said consuming units for the time interval using both historical and weather forecast data. The method also includes determining the maximum load capacity of each segment in the utility grid for the time interval, the segments being the power lines making up the grid, and assigning the maximum load capacity of each segment to be a constraint therefor for the time interval and computing whether there are paths from each of the generating units to each of the consuming units for which the maximum load capacities will not be exceeded. If such paths exist, the method further concludes with setting the utility grid to the paths; and transmitting electric power over the paths for the time interval. If such paths do not exist from each generating unit to each consuming unit, the method instead continues by reducing the maximum capacities of the generating units; and recomputing whether there are paths from each of the generating units to each of the consuming units as many times as is necessary to find paths satisfying the constraints, each time further reducing the maximum capacities of the generating units.

Claims

1. A method for a utility grid having a network of interconnecting segments of power lines connecting generating units to consuming units, each of said interconnecting segments-having a maximum electric power capacity, said method comprising:

computing a probable electric power output for each of a plurality of generating units, each of said plurality of generating units having a maximum electric power generating capacity, said plurality of generating units including at least one generating unit whose electric power output depends on the weather, and a respective confidence level associated with each of said plurality of generating units, for a given time interval;

computing a probable electrical power load for each of a plurality of consuming units, each of said plurality of consuming units having a maximum allowed electric power load, and a respective confidence level associated with each of said plurality of consuming units, for said time interval; and

determining whether there are paths having one or more segments of power lines from each of said generating units to each of said consuming units, said paths satisfying the following constraints;

a) the actual output of each of said generating units is less than or equal to the probable output thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof for said time interval; and

c) the actual load capacity for each of said segments is less than or equal to the maximum load capacity thereof for said time interval; and

when there are paths satisfying said constraints;

setting the utility grid to said paths; and

transmitting electric power over said paths for said time interval.

2. The method as claimed in claim 1, further comprising, when there is no path between one of said generating units and one of said consuming units satisfying said constraints:

reducing the maximum capacity of at least one of said generating units; and

re-computing whether there are paths from each of said generating units to each of said consuming units as many times as is necessary to find paths satisfying said constraints, each time reducing the capacity of at least one of said generating units.

3. The method as claimed in claim 1, wherein a length of said given time interval is between five and ten hours.

4. The method as claimed in claim 1, wherein said maximum capacities of all generating units, said maximum load requirements of all consuming units, and said maximum capacities of all segments all have latencies, said latencies being time periods over which predictions relating thereto by stochastic forecasting may be made, and wherein a length of said given time interval is longer than the longest of said latencies.

5. The method as claimed in claim 1, wherein the probable output for each generating unit whose electric power output depends on the weather is determined using a stochastic model based on weather prediction.

6. The method as claimed in claim 5, wherein the stochastic model uses historical data to determine the amount of power that will be generated along with a confidence level for said time interval.

7. The method as claimed in claim 1, wherein the probable load for each of said consuming units is determined using a stochastic model based on weather prediction.

8. The method as claimed in claim 7, wherein the stochastic model uses historical data and current input, including cost of power, time of day, season, and day of the week, to determine the probable load along with a confidence level for said time interval.

9. The method as claimed in claim 1, wherein the maximum load capacity of each segment in said utility grid for said time interval is a rated capacity modified based on the physical condition of the segment as well as on weather conditions for said time interval.

10. The method as claimed in claim 1, wherein said paths from each of said generating units to each of said consuming units are computed using a greedy algorithm.

11. The method as claimed in claim 1, wherein said constraints for the actual output of each of said generating units and for the actual load of each of said consuming units are modified as follows:

a) the actual output of each of said generating units is less than or equal to the probable output thereof plus the confidence level thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof minus the confidence level thereof for said time interval.

12. An apparatus comprising:

one or more processors; and

one or more memories including computer program code, the one or more memories and the computer program code configured, with the one or more processors, to cause the apparatus to perform the following for a utility grid having a network of interconnecting segments of power lines connecting generating units to consuming units, each of said interconnecting segments-having a maximum electric power capacity;

computing a probable electric power output for each of a plurality of generating units, each of said plurality of generating units having a maximum electric power generating capacity, said plurality of generating units including at least one generating unit whose electric power output depends on the weather, and a respective confidence level associated with each of said plurality of generating units, for a given time interval;

computing a probable electrical power load for each of a plurality of consuming units, each of said plurality of consuming units having a maximum allowed electric power load, and a respective confidence level associated with each of said plurality of consuming units, for said time interval; and

determining whether there are paths having one or more segments of power lines from each of said generating units to each of said consuming units, said paths satisfying the following constraints;

a) the actual output of each of said generating units is less than or equal to the probable output thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof for said time interval; and

c) the actual load capacity for each of said segments is less than or equal to the maximum load capacity thereof for said time interval; and

when there are paths satisfying said constraints;

setting the utility grid to said paths; and

transmitting electric power over said paths for said time interval.

13. The apparatus as claimed in claim 12, wherein said one or more memories and the computer program code are further configured, with the one or more processors, to cause the apparatus to perform, when there is no path between one of said generating units and one of said consuming units satisfying said constraints:

reducing the maximum capacity of at least one of said generating units; and

re-computing whether there are paths from each of said generating units to each of said consuming units as many times as is necessary to find paths satisfying said constraints, each time reducing the capacity of at least one of said generating units.

14. The apparatus as claimed in claim 12, wherein a length of said given time interval is between five and ten hours.

15. The apparatus as claimed in claim 12, wherein said maximum capacities of all generating units, said maximum load requirements of all consuming units, and said maximum capacities of all segments all have latencies, said latencies being time periods over which predictions relating thereto by stochastic forecasting may be made, and wherein a length of said given time interval is longer than the longest of said latencies.

16. The apparatus as claimed in claim 12, wherein the probable output for each generating unit whose electric power output depends on the weather is determined using a stochastic model based on weather prediction.

17. The apparatus as claimed in claim 16, wherein the stochastic model uses historical data to determine the amount of power that will be generated along with a confidence level for said time interval.

18. The apparatus as claimed in claim 12, wherein the probable load for each of said consuming units is determined using a stochastic model based on weather prediction.

19. The apparatus as claimed in claim 18, wherein the stochastic model uses historical data and current input, including cost of power, time of day, season, and day of the week, to determine the probable load along with a confidence level for said time interval.

20. The apparatus as claimed in claim 12, wherein the maximum load capacity of each segment in said utility grid for said time interval is a rated capacity modified based on the physical condition of the segment as well as on weather conditions for said time interval.

21. The apparatus as claimed in claim 12, wherein said paths from each of said generating units to each of said consuming units are computed using a greedy algorithm.

22. The apparatus as claimed in claim 12, wherein said constraints for the actual output of each of said generating units and for the actual load of each of said consuming units are modified as follows:

a) the actual output of each of said generating units is less than or equal to the probable output thereof plus the confidence level thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof minus the confidence level thereof for said time interval.

23. A computer program product comprising a computer-readable storage medium bearing computer program code embodied therein for use with a computer, the computer program code comprising code for performing at least the following for a utility grid having a network of interconnecting segments of power lines connecting generating units to consuming units, each of said interconnecting segments-having a maximum electric power capacity:

computing a probable electric power output for each of a plurality of generating units, each of said plurality of generating units having a maximum electric power generating capacity, said plurality of generating units including at least one generating unit whose electric power output depends on the weather, and a respective confidence level associated with each of said plurality of generating units, for a given time interval;

computing a probable electrical power load for each of a plurality of consuming units, each of said plurality of consuming units having a maximum allowed electric power load, and a respective confidence level associated with each of said plurality of consuming units, for said time interval; and

determining whether there are paths having one or more segments of power lines from each of said generating units to each of said consuming units, said paths satisfying the following constraints;

a) the actual output of each of said generating units is less than or equal to the probable output thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof for said time interval; and

c) the actual load capacity for each of said segments is less than or equal to the maximum load capacity thereof for said time interval; and

when there are paths satisfying said constraints;

setting the utility grid to said paths; and

transmitting electric power over said paths for said time interval.

24. The computer program product as claimed in claim 23, wherein the computer program code further comprises code for performing, when there is no path between one of said generating units and one of said consuming units satisfying said constraints:

reducing the maximum capacity of at least one of said generating units; and

re-computing whether there are paths from each of said generating units to each of said consuming units as many times as is necessary to find paths satisfying said constraints, each time reducing the capacity of at least one of said generating units.

25. The computer program product as claimed in claim 23, wherein a length of said given time interval is between five and ten hours.

26. The computer program product as claimed in claim 23, wherein said maximum capacities of all generating units, said maximum load requirements of all consuming units, and said maximum capacities of all segments all have latencies, said latencies being time periods over which predictions relating thereto by stochastic forecasting may be made, and wherein a length of said given time interval is longer than the longest of said latencies.

27. The computer program product as claimed in claim 23, wherein the probable output for each generating unit whose electric power output depends on the weather is determined using a stochastic model based on weather prediction.

28. The computer program product as claimed in claim 27, wherein the stochastic model uses historical data to determine the amount of power that will be generated along with a confidence level for said time interval.

29. The computer program product as claimed in claim 23, wherein the probable load for each of said consuming units is determined using a stochastic model based on weather prediction.

30. The computer program product as claimed in claim 29, wherein the stochastic model uses historical data and current input, including cost of power, time of day, season, and day of the week, to determine the probable load along with a confidence level for said time interval.

31. The computer program product as claimed in claim 23, wherein the maximum load capacity of each segment in said utility grid for said time interval is a rated capacity modified based on the physical condition of the segment as well as on weather conditions for said time interval.

32. The computer program product as claimed in claim 23, wherein said paths from each of said generating units to each of said consuming units are computed using a greedy algorithm.

33. The computer program product as claimed in claim 23, wherein said constraints for the actual output of each of said generating units and for the actual load of each of said consuming units are modified as follows:

a) the actual output of each of said generating units is less than or equal to the probable output thereof plus the confidence level thereof for said time interval;

b) the actual load of each of said consuming units is less than or equal to the probable load thereof minus the confidence level thereof for said time interval.