An Optimization-Based Line-Wise Approach for Accurate Radial Distribution System State Estimation

Abstract

Distribution system state estimation (DSSE) is an essential tool for the effective operation and management of modern distribution systems. A common challenge in DSSE is ensuring accurate estimates despite limited real-time measurements and high pseudo-measurement errors. This paper presents a novel line-wise state estimator (LW-SE) for radial distribution systems, leveraging conic quadratic optimization to transform the non-convex state estimation problem into a convex one. This transformation enhances the accuracy of the state estimation process. Unlike traditional methods, the LW-SE formulation uses line impedances rather than admittances, addressing issues associated with low-impedance branches and leading to more stable power flow representations. Furthermore, the method accommodates diverse types of measurements without requiring paired active and reactive power measurements, their equivalent forms, or phase angle measurements as inputs—while still enabling accurate phase angle estimation. Results of case studies and comparisons with traditional state estimators (T-SE) demonstrate the effectiveness of the LW-SE with accuracy improvement ranging from 60% to 82% in scenarios with low availability of real-time measurements and high errors in pseudo-measurement. In scenarios involving gross measurement errors, the LW-SE consistently delivered lower MAPEs than the weighted least squares (WLS) and weighted least absolute value (WLAV) state estimators, while maintaining computational efficiency. These findings underscore the LW-SE's suitability for modern distribution system applications.