Abstract: Based On the study about the application of Fault Tree Analysis and Bidirectiongal Associative Memory network in fault diagnosis, a method of fault diagnosis based on the fuse of FTA and BAM is proposed.In a system, all the knowledge on the happening of top events is stored in Fault tree, so all of the fault modes are obtained by using FTA, and then the learning sample of BAM are summarized which are the corresponding relations between the fault phenomena (the states of the assembled monitoring points) and the happening or not of the bottom events in fault tree.The results of diagnosis are associated parallel by the associative memory matrix, thus expanding the general ability of fault diagnosis.A simulation results showed that the approach is valid to the fault diagnosis of systems.