Topology error detection and identification in network analysis

Errors in power system topology are difficult to handle with classical methods based on state estimation. This paper proposes a pre-processing method for detecting and identifying topology errors and bad measurements before a state estimator solution. Two algorithms are described. These are used at the substation level or at the substation voltage level.The first algorithm is more suited to systems (substations) where no switch flow measurements are available. Therefore, it does not require any additional measurements other than the ones currently used by the state estimator. The topology error detection uses a bus-branch model. The model relies on the connectivity information and on the impedance values of the non-zero impedance branches.The second algorithm is more suited to systems (substations) where some switch flow measurements are available. Even if these are not required by the algorithm, using them add redundancy and provides a more reliable solution. The topology error detection uses a bus section—switch device model. The model uses only branch connectivity.The first algorithm provides better results when no switch flow measurements are available. The second one is superior when some switch flow measurements are available. These algorithms complement each other and they can be use together to cover a much broader scope of the topology error detection and identification problem.