In this paper, sequences of events registered in power distribution systems are analysed using pattern sequence discovery algorithms. The events considered in the study are basically voltage sags generated by homopolar faults and registered by power quality monitors installed in the secondary of transformers in distribution substations. The events registered in a measuring point have associated the time of occurrence, and the list of increasing-time ordered events corresponds to a sequence. The aim is to discover existing temporal patterns, or episodes, that occur sufficiently often along that sequence. Sequence pattern discovery algorithms are used to discover those frequent episodes. Then, frequent episodes are used to obtain association rules that describe relationships between time spans of successive events. The method has been tested with data from gathered in different substations.