State estimation of electrical power networks using semidefinite relaxation

Patent metadata

Field	Value
Publication number	US-9863985-B2
Application number	US-201313862139-A
Country	US
Kind code	B2
Filing date	Apr 12, 2013
Priority date	Apr 13, 2012
Publication date	Jan 9, 2018
Grant date	Jan 9, 2018

Abstract

A semidefinite (SDR) programming formulation for state estimation (SE) of nonlinear AC power systems is described. The techniques make use of convex semidefinite relaxation of the original problem to render the process efficiently solvable. In addition, robust techniques are described that are resilient to outlying measurements and/or adversarial cyber-attacks. Further, techniques for SDR-based SE are described in which local control areas solve the centralized SDP-based SE problem in a distributed fashion.

First claim

The invention claimed is:

1. A method for estimating a state for each of a plurality of alternating current (AC) electrical power buses within a power grid, the method comprising:
receiving, with an energy management system for the power grid, measurements of an electrical characteristic from a plurality of measurement units (MUs) positioned at a subset of the power buses within the power grid;
processing the measurements of the electrical characteristic with the energy management system to compute an estimate of the electrical characteristic currently occurring at each of the power buses within the power grid, the estimates of the electrical characteristic being nonlinearly related to the measurements of the electrical characteristic, wherein processing the measurements of the electrical characteristic comprises:
applying semidefinite relaxation to compute a semidefinite programming model for the power buses within the power grid that linearly relates the estimates for the electrical characteristic to the measurements of the electrical characteristic; and
iteratively processing the measurements of the electrical characteristic in accordance with the semidefinite programming model to compute a convex minimization as a solution for the estimates of the electrical characteristic; and

controlling operation of one or more of the power buses based on the computed estimates.

2. The method of claim 1 , wherein the measurements of the electrical characteristic comprises measurements of one or more of a current phasor, a voltage phasor or a power phasor.

3. The method of claim 1 , wherein the measurements of the electrical characteristic comprises a complex voltage at each of the power buses including: (1) a real power injection into the power grid at the power bus, (2) a real power flow to the power grid at the power bus, and (3) a voltage magnitude at the power bus.

4. The method of claim 1 , wherein the measurements of the electrical characteristic are received from phasor measurement units (PMUs).

5. The method of claim 1 , wherein applying semidefinite relaxation to compute a semidefinite programming model for the power buses comprises defining the set of estimates of one or more voltage phasors, {circumflex over (v)}, such that

v
^

:=

arg
⁢

⁢

min
v

⁢

∑

l
=
1

L

⁢

w
l

⁡

[

z
l

-

h
l

⁡

(
v
)

]

2

.

6. The method of claim 1 , further comprising generating a report that specifies power consumption at each of the power buses within the power grid based on the computed estimates.

7. The method of claim 6 , further comprising updating an accounting system to charge respective operators of each of the power buses based on the computed estimates.

8. The method of claim 1 , further comprising generating an alert based on the computed estimates.

9. The method of claim 1 ,
wherein the semidefinite programming model includes a state vector indicating a trustworthiness for each of the measurements, and
wherein processing the measurements comprises computing the solution by eliminating any outliers of meter measurements based on by trustworthiness of each of the measurements.

10. The method of claim 9 , wherein the state vector identifies outliers including faulty measurements, bad data due to communication error and data that has been compromised due to attack.

11. The method of claim 9 , wherein the semidefinite programming model includes a configurable parameter to control a required level of trustworthiness for the measurements.

12. The method of claim 1 ,
wherein the energy management system is one of a plurality of energy management systems associated with respective power grids, each of the energy management systems computing estimates of the electrical characteristic for electrical power buses within the respective power grid,
the method further comprising:
communicating the estimates from each of the energy management systems to a central energy management system; and
iteratively processing the estimates from the energy management systems in accordance with a semidefinite programming model to compute a global solution for the estimates of the electrical characteristics for the entire plurality of power grids.

13. The method of claim 1 ,
wherein the energy management system is one of a plurality of energy management systems associated with respective power sub-grids that are interconnected to form the power grid, each of the power sub-grids including a subset of the electrical busses and representing a control area for the corresponding energy management system, and
wherein processing the measurements comprises, with each of the energy management systems, locally computing the state estimates of the electrical characteristic for the subset of the electrical power buses within the respective control area associated with the respective power sub-grid.

14. The method of claim 13 , further comprising:
expanding the subset of the electrical power buses for a first one of the control areas to include one of the electrical busses of a second one of the control areas when a transmission line connects the first one of the control areas to the electrical bus of the second control area and the first one of the control areas includes a sensor that provides a measurement associated with the transmission line; and
communicating, during the computation of the solution for the state estimates, intermediate results for the state estimates from the second one of the control areas to the first one of the control areas for the electrical bus to which the transmission line is connected.

15. The method of claim 13 , further comprising,
for each of the control areas, identifying each transmission line that connects the control area to an electrical bus within a different one of the control areas and includes a sensor that provides a measurement associa

Assignees

• Univ Minnesota

Inventors

• Giannakis Georgios B
• Zhu Hao

Classifications

• H02J2103/30

  Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks · CPC title

• H02J13/1331

  using wireless data transmission · CPC title

• H02J13/12

  Monitoring network conditions, e.g. electrical magnitudes or operational status · CPC title

• G06Q20/145

  Payments according to the detected use or quantity · CPC title

• G01R21/133 Primary

  by using digital technique · CPC title