Computational Enhancement of Sparse Tableau via Block LU Factorization for Power Flow Studies

Abstract

Recent work has proposed alternatives to avoid the limitations that are inherent in Y-bus-based formulations for power flow problems. Alternatives based on the Sparse Tableau Formulation (STF) provide conceptual benefits relative to Y(bus )methods. This letter focuses on specific computational approaches tailored to the features of STF, that allow it to match the computational speed of well-established Y-bus-based methods. In particular, it presents enhanced Newton Raphson (NR) algorithms exploiting the block LU factorization to improve the computational performance of STF. These methods are compared with the classic Y-bus-based solution algorithm using several power system test networks. Com-putational case studies demonstrate the significant computational improvement for the STF-based power flow solution.