This paper investigates the use of topology identification algorithms for detection of network configuration changes in Active Distribution Networks (ADNs). In ADNs with high penetrations of Distributed Generation (DG) and microgrids (μGs), network topology identification is more complex, since there are more switching operations and a greater need for the State Estimator (SE) to be robust to missing or incorrect switch statuses. This paper develops and tests SE algorithms with an event-triggered topology identification stage designed to identify network configuration changes, including the connection/disconnection of μGs. The methodology is demonstrated using recorded MV distribution system measurement data, and its performance is investigated in cases where the input measurement data quality and redundancy is low. The performance of two different approaches to the topology identification problem, the Recursive Bayesian Approach (RBA) and the Generalised SE approach, are compared.