Normalized Deleted Residual Test for Identifying Interacting Bad Data in Power System State Estimation

The Largest Normalized Residual Test (LNRT) has been widely utilized in commercial Power System State Estimation (PSSE) software for bad data identification. LNRT proved effective dealing with single as well multiple non-interacting and interacting but non-conforming data. However, it is known a long time that when two are both conforming, i.e. their errors agreement, the may fail to identify either one. Moreover, shown recently even can cause failure of LNRT. Drawing on sensitivity analysis linear regression, we develop normalized deleted residuals suspected measurements so agreement measurement broken. Therefore, will be able actual point. Furthermore, case AC PSSE, method does not require calculation new hat matrix from set. This makes computationally cost-effective. Simulation results identifying different conforming proves proposed enhance effectiveness