In generalized state estimation， to identify the misinformation of the connected region formed by the closed breakers， the concept of virtual breakers is used to build the minimum feature network based on the generalized network that does not affect the quality of identification. In addition to analog measurements， zero-voltage-drop for virtual closed switch as equality constraint is added to formulate the model of weighted least squares with equality constraints. The relation between residual and Lagrange multiplier from the probability standpoint is discussed alongside the approach to identifying bad data. Furthermore， it can be seen that the identification of the bad analog data is distinguished from that of the bad breaker status under normal conditions. Finally， the validity of the model and algorithm is verified with a three-node simple system and the data from Yantai grid of Shandong Province.