skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids

Abstract

We present an augmented matrix approach to update the solution to a linear system of equations when the coefficient matrix is modified by a few elements within a principal submatrix. This problem arises in the dynamic security analysis of a power grid, where operators need to perform $N-x$ contingency analysis, i.e., determine the state of the system when up to $x$ links from $N$ fail. Our algorithms augment the coefficient matrix to account for the changes in it, and then compute the solution to the augmented system without refactoring the modified matrix. We provide two algorithms, a direct method, and a hybrid direct-iterative method for solving the augmented system. We also exploit the sparsity of the matrices and vectors to accelerate the overall computation. Our algorithms are compared on three power grids with PARDISO, a parallel direct solver, and CHOLMOD, a direct solver with the ability to modify the Cholesky factors of the coefficient matrix. We show that our augmented algorithms outperform PARDISO (by two orders of magnitude), and CHOLMOD (by a factor of up to 5). Further, our algorithms scale better than CHOLMOD as the number of elements updated increases. The solutions are computed with high accuracy. Our algorithms are capable of computing $N-x$ contingency analysis on a $778K$ bus grid, updating a solution with $x=20$ elements in $$1.6 \times 10^{-2}$$ seconds on an Intel Xeon processor.

Authors:
; ; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1416692
Report Number(s):
PNNL-SA-119762
KJ0401000
DOE Contract Number:
AC05-76RL01830
Resource Type:
Journal Article
Resource Relation:
Journal Name: SIAM Journal on Scientific Computing, 39(5):S809 -- S827
Country of Publication:
United States
Language:
English

Citation Formats

Yeung, Yu-Hong, Pothen, Alex, Halappanavar, Mahantesh, and Huang, Zhenyu. AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids. United States: N. p., 2017. Web. doi:10.1137/16M1082755.
Yeung, Yu-Hong, Pothen, Alex, Halappanavar, Mahantesh, & Huang, Zhenyu. AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids. United States. doi:10.1137/16M1082755.
Yeung, Yu-Hong, Pothen, Alex, Halappanavar, Mahantesh, and Huang, Zhenyu. 2017. "AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids". United States. doi:10.1137/16M1082755.
@article{osti_1416692,
title = {AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids},
author = {Yeung, Yu-Hong and Pothen, Alex and Halappanavar, Mahantesh and Huang, Zhenyu},
abstractNote = {We present an augmented matrix approach to update the solution to a linear system of equations when the coefficient matrix is modified by a few elements within a principal submatrix. This problem arises in the dynamic security analysis of a power grid, where operators need to perform $N-x$ contingency analysis, i.e., determine the state of the system when up to $x$ links from $N$ fail. Our algorithms augment the coefficient matrix to account for the changes in it, and then compute the solution to the augmented system without refactoring the modified matrix. We provide two algorithms, a direct method, and a hybrid direct-iterative method for solving the augmented system. We also exploit the sparsity of the matrices and vectors to accelerate the overall computation. Our algorithms are compared on three power grids with PARDISO, a parallel direct solver, and CHOLMOD, a direct solver with the ability to modify the Cholesky factors of the coefficient matrix. We show that our augmented algorithms outperform PARDISO (by two orders of magnitude), and CHOLMOD (by a factor of up to 5). Further, our algorithms scale better than CHOLMOD as the number of elements updated increases. The solutions are computed with high accuracy. Our algorithms are capable of computing $N-x$ contingency analysis on a $778K$ bus grid, updating a solution with $x=20$ elements in $1.6 \times 10^{-2}$ seconds on an Intel Xeon processor.},
doi = {10.1137/16M1082755},
journal = {SIAM Journal on Scientific Computing, 39(5):S809 -- S827},
number = ,
volume = ,
place = {United States},
year = 2017,
month =
}
  • Cited by 41
  • An algorithm is presented for solving a block-tridiagonal, linear matrix equation containing a singular submatrix whose inverse cannot be obtained. In applying the algorithm to distillation calculations, if the temperature of a specified plate is given, the residual function of the plate must be replaced with another residual function, producing a singular Jacobian submatrix with a zero row; if the feed enthalpy is unknown, a variable, viz., the vapor flow rate at the top of the column, must be replaced with the enthalpy of the feed, producing a singular Jacobian submatrix with a zero column. The effectiveness of the algorithmmore » is demonstrated for the two cases by numerical examples.« less
  • Curved-wave multiple-scattering contributions to XAFS (x-ray-absorption fine structure) are calculated with use of an efficient formalism similar to that based on the plane-wave approximation, but with scattering amplitudes {ital f}({theta}) replaced by distance-dependent scattering matrices'' {ital F}{sub {lambda},{lambda}{prime}}({rho},{rho}{prime}). Here {rho}={ital k}{bold R}, {ital k} being the photoelectron wave number and {bold R} is a bond vector, while the matrix indices {lambda}=({mu},{nu}) represent terms in a convergent expansion that generalizes the small-atom approximation. This approach is based on an exact, separable representation of the free propagator (or translation operator) matrix elements, {ital G}{sub {ital L},{ital L{prime}}}({ital k}{bold R}), in an angularmore » momentum {ital L}=({ital l},{ital m}) and site basis. The method yields accurate curved-wave contributions for arbitrarily high-order multiple-scattering paths at all positive energies, including the near-edge region. Results are nearly converged when the intermediate {lambda} summations are truncated at just six terms, i.e., (6{times}6) matrices. The lowest-order (1{times}1) matrix {ital F}{sub 00,00} is the effective, curved-wave scattering amplitude, {ital f}({rho},{rho}{prime},{theta}), and yields a multiple-scattering expansion equivalent to the point-scattering approximation. Formulas for multiple-scattering contributions to XAFS and photoelectron diffraction are presented, and the method is illustrated with results for selected multiple-scattering paths in fcc Cu.« less
  • The analysis of homonuclear conjugated systems is extended to include the interactions between non-neighbor atoms and the variation in the interactions between neighbor atoms. Both kinds of interactions are included as perturbation effects on the tight-binding approximation. The formalism developed for the latter is not complicated and the interpretations remain unchanged. Application to benzene and naphthalene illustrates the approach. (auth)