System and method for phase balancing in a power distribution system

• US 9,041,246 B2
• Filed: 09/29/2010
• Issued: 05/26/2015
• Est. Priority Date: 09/29/2010
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Black, Jason Wayne, Tinnium, Kannan Natarajan, Larson, Ronald Ray, Wang, Xiaofeng, Johal, Harjeet
• Primary Examiner(s): Barnie, Rexford
• Assistant Examiner(s): Mourad, Rasem
• Application Number:
  US12/893,864
• Publication Number:
  US 20120074779A1
• Time in Patent Office: 1,700 Days
• Field of Search: 307/11, 307/14, 307/28, 700/291, 700/295, 700/298
• US Class Current: 307/14
• CPC Class Codes: H02J 3/003 Load forecast, e.g. methods or systems for forecasting future load demand
• CPC Class Codes: H02J 3/26 Arrangements for eliminating or reducing asymmetry in polyphase networks
• CPC Class Codes: Y02E 40/50 Arrangements for eliminating or reducing asymmetry in polyphase networks
• CPC Class Codes: Y04S 10/50 Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

A phase balancing system includes a load forecasting module, a phase unbalance identification module and a demand response module. The load forecasting module determines a load forecast for the distribution system for the period of interest and the phase unbalance identification module determines voltage unbalance on the distribution system for the period of interest. The demand response module estimates an available demand response on the distribution system for the period of interest and allocates an optimized demand response from the available demand response to minimize the voltage unbalance on the distribution system for the period of interest.

Claims

1. A phase balancing system for a distribution system comprising:

a load forecasting module including a combination of software and hardware elements for determining a load forecast for the distribution system for a period of interest;

a phase unbalance identification module including a combination of software and hardware elements for determining voltage unbalance on the distribution system for the period of interest based on the load forecast;

a demand response module including a combination of software and hardware elements for receiving the voltage unbalance from the phase unbalance identification module and for estimating an available demand response on the distribution system for the period of interest;

wherein the demand response module is further configured for allocating an optimized demand response from the available demand response and controlling loads in the distribution system to minimize the voltage unbalance on the distribution system for the period of interest.

2. The phase balancing system of claim 1, wherein the load forecasting module comprises at least one of a regression algorithm, a time series algorithm, a neutral network algorithm, a fuzzy logic algorithm or a statistical learning algorithm.

3. The phase balancing system of claim 2, wherein the load forecast is determined based on time, weather conditions, customer types, distribution system conditions and historical data.

4. The phase balancing system of claim 3, wherein distribution system conditions include topology of the distribution system, location of customers along the network, and information on control devices.

5. The phase balancing system of claim 4, wherein information on control devices includes information on voltage regulators and capacitor banks.

6. The phase balancing system of claim 1, wherein phase unbalance identification module comprises a load flow algorithm to determine voltage information at various nodes in the distribution system.

7. The phase balancing system of claim 6, wherein the voltage information comprises a voltage angle and a voltage magnitude information.

8. The phase balancing system of claim 6, wherein the load flow algorithm comprises a William Kersting'"'"'s backward/forward sweep algorithm.

9. The phase balancing system of claim 1, wherein the demand response module comprises a critical peak pricing (CPP) program, a variable peak pricing (VPP) program or a direct load control (DLC) program.

10. The phase balancing system of claim 1, wherein the period of interest comprises a number of time steps.

11. The phase balancing system of claim 1, wherein the period of interest comprises an hour, a day or a week.

12. The phase balancing system of claim 1, wherein allocating an optimized demand response comprises utilizing one of a dynamic programming, an evolutionary programming, a neutral network and fuzzy system, a heuristic technique, a genetic algorithm, or an interior point method to identify the optimized demand response.

13. A method for phase balancing in a distribution system comprising:

determining a load forecast for the distribution system for a period of interest;

determining a voltage unbalance on the distribution system for the period of interest based on the load forecast;

estimating an available demand response on the distribution system for the period of interest; and

allocating an optimized demand response from the available demand response and controlling loads in the distribution system to minimize the determined voltage unbalance on the distribution system.

14. The method of claim 13, wherein the load forecast is determined based on time, weather conditions, customer types, distribution system conditions and historical data.

15. The method of claim 13, wherein determining the voltage unbalance comprises determining voltage information on each of the node in the distribution system by applying a load flow algorithm on the load forecast.

16. The method of claim 15, wherein the voltage information comprises a voltage angle and a voltage magnitude information.

17. The method of claim 15, wherein the load flow algorithm comprises a William Kersting'"'"'s backward/forward sweep algorithm.

18. The method of claim 13, wherein determining the available demand response comprises utilizing at least one of a critical peak pricing (CPP) program, a variable peak pricing (VPP) program or a direct load control (DLC) program.

19. The method of claim 13, wherein allocating an optimized demand response comprises utilizing one of a dynamic programming, an evolutionary programming, a neutral network and fuzzy system, a genetic algorithm, or an interior point method to identify the optimized demand response.

20. A non-transitory computer-readable medium comprising computer-readable instructions of a computer program that, when executed by a processor, cause the processor to perform a method for phase balancing in a distribution system comprising:

determining a load forecast for the distribution system for a period of interest;

determining a voltage unbalance on the distribution system for the period of interest based on the load forecast;

estimating an available demand response on the distribution system for the period of interest; and

allocating an optimized demand response from the available demand response and controlling loads in the distribution system to minimize the determined voltage unbalance on the distribution system.

21. The non-transitory computer-readable medium of claim 20, wherein load forecast is determined based on time, weather conditions, customer types, distribution system conditions and historical data.

22. The non-transitory computer-readable medium of claim 20, wherein determining the voltage unbalance comprises determining voltage information on each of the node in the distribution system by applying a load flow algorithm on the load forecast.

23. The non-transitory computer-readable medium of claim 22, wherein the voltage information comprises a voltage angle and a voltage magnitude information.

24. The non-transitory computer-readable medium of claim 22, wherein the load flow algorithm comprises a William Kersting'"'"'s backward/forward sweep algorithm.

25. The non-transitory computer-readable medium of claim 20, wherein determining the available demand response comprises utilizing at least one of a critical peak pricing (CPP) program, a variable peak pricing (VPP) program or a direct load control (DLC) program.