Voltage stability assessment, control and probabilistic power flow based on multi-dimensional holomorphic embedding techniques
A multi-dimensional holomorphic embedding method for voltage control of an AC power system includes embedding multiple independent symbolic variables representing multiple control elements of the AC power system into AC power flow equations that describe the AC power system, analytically solving voltages for targeted buses of the AC power system in a form of multivariate power series or multivariable Padé approximants about the multiple independent symbolic variables such that coefficients of the multivariate power series or multivariable Padé approximants are obtained non-iteratively, and jointly adjusting the multiple control elements according to the multivariate power series or multivariable Padé approximants to control voltages of the targeted buses.
1. A multi-dimensional holomorphic embedding method for voltage control of an AC power system:
by one or more processors,
embedding multiple independent symbolic variables representing multiple control elements of the AC power system into AC power flow equations that describe the AC power system,
analytically solving voltages for targeted buses of the AC power system in a form of multivariate power series about the multiple independent symbolic variables or multivariable Padé approximants about the multiple independent symbolic variables such that coefficients of the multivariate power series or multivariable Padé approximants are obtained non-iteratively, and
jointly adjusting the multiple control elements according to the multivariate power series or multivariable Padé approximants to control voltages of the targeted buses.
2. The multi-dimensional holomorphic embedding method of claim 1 further comprising performing probabilistic power flow analysis on status of the AC power system based on the multivariate power series.
3. The multi-dimensional holomorphic embedding method of claim 2 , wherein the multiple independent symbolic variables are input random variables describing power injections to the AC power system, further comprising deriving cumulants of the voltages or line power of the AC power system by a generalized cumulant method.
4. The multi-dimensional holomorphic embedding method of claim 3 further comprising obtaining distributions of the voltages or line power of the AC power system based on the cumulants by Gram-Charlier method, Edgeworth method, or Cornish-Fisher method.
5. The multi-dimensional holomorphic embedding method of claim 1 , wherein the multiple control elements include generators, shunt capacitors, shunt reactors, static synchronous compensators, or static VAR compensators.
6. The multi-dimensional holomorphic embedding method of claim 1 , wherein the analytically solving includes identifying a physical germ solution describing initial physical conditions of the AC power system.
7. The multi-dimensional holomorphic embedding method of claim 1 , wherein the analytically solving includes expressing variables of the AC power flow equations in the form of the multivariate power series.
8. The multi-dimensional holomorphic embedding method of claim 1 , wherein the analytically solving includes recursively obtaining the coefficients of the multivariate power series.
9. The multi-dimensional holomorphic embedding method of claim 1 , wherein the analytically solving includes transforming the multivariate power series into multivariable Padé approximants to increase a radius of convergence of analytical solutions of all variables in the AC power flow equations.
10. A method for voltage stability assessment and control for AC power system load areas to prevent voltage collapse, the method comprising:
by one or more processors,
embedding multiple independent symbolic variables representing a next-period load increase for all load buses of the AC power system load areas into AC power flow equations,
analytically expressing power-voltage curves for the load buses in a form of multivariate power series about the multiple independent symbolic variables or Padé approximants about the multiple independent symbolic variables to identify a voltage stability margin for each of the load buses, and
controlling load consumption of the load buses according to minima of the voltage stability margins to prevent voltage collapse on the load buses.
11. The method of claim 10 , wherein the Padé approximants are adaptive two-stage Padé approximants.
12. A method for distributing reactive power among a plurality of reactive power resources of an AC power system to remotely control voltages of the AC power system, the method comprising:
by one or more processors,
modifying bus-types of buses of AC power flow equations describing remote voltage control functionality of the AC power system such that each of the bus-types corresponds to an embedding method of the AC power flow equations,
embedding participation factors, that define how the reactive power is to be distributed among the plurality of reactive power resources, into the AC power flow equations,
solving the AC power flow equations in a form of multivariate power series about multiple independent symbolic variables or Padé approximants about multiple independent symbolic variables such that coefficients of the power series or Padé approximants are obtained non-iteratively, wherein references of the plurality of reactive power resources are calculated from the coefficients, and
controlling outputs of the plurality of reactive power resources in an operational environment according to the references to maintain bus voltage magnitudes of selected ones of the buses.
13. The method of claim 12 , wherein the plurality of reactive power resources includes condensers, generators, static synchronous compensators, static VAR compensators, or voltage source converters-high voltage direct current.