Bad Data Suppression in Power System State Estimation with a Variable Quadratic-Constant Criterion

This paper reports numerical results on tests of a new bad data suppression strategy for power system state estimation. A transformation-decoupled state estimator [1] is used in conjunction with a variable quadratic-constant criterion whose break even point is designed to track the upper bound of normalized residues of good measurements. Numerical simulations with a 6-bus test system and the IEEE 30-bus system show that the new bad data suppression strategy, is very effective for different systems under many uncertainties in the sizes and numbers of bad data with a moderate measurement redundancy.