Hybrid Symbolic-Numeric Framework for Power System Modeling and Analysis

Abstract

With the recent proliferation of open-source packages for computing, power system differential-algebraic equation (DAE) modeling and simulation are being revisited to reduce the programming efforts. Existing open-source tools require manual efforts to develop code for numerical equations, sparse Jacobians, and discontinuous components. This paper proposes a hybrid symbolic-numeric framework, exemplified by an open-source Python-based library ANDES, which consists of a symbolic layer for descriptive modeling and a numeric layer for vector-based numerical computation. This method enables the implementation of DAE models by mixing and matching modeling components, through which models are described. In the framework, a rich set of discontinuous components and standard transfer function blocks are provided besides essential modeling elements for rapid modeling. ANDES can automatically generate robust and fast numerical simulation code, as well as and high-quality documentation. Case studies present a) two implementations of turbine governor model TGOV1, b) power flow computation time break down for MATPOWER systems, c) validation of time-domain simulation with commercial software using three test systems with a variety of models, and d) the full eigenvalue analysis for Kundur's system. Validation shows that ANDES closely matches the commercial tool DSATools for power flow, time-domain simulation, and eigenvalue analysis.