A research on the fault diagnosis technique of automatic shift control system in steady state conditions

Fault detection and diagnosis of automatic shift control system (ASCS) is fulfilled by using principal component analysis technique. Based on the state variable characteristics of ASCS in steady state conditions, the principal component model for ASCS is established with faultless historical data. According to the feature of multivariate normal distribution of data, a comprehensive monitoring indicator O_Index is adopted instead of commonly used T² and Q statistics. When fault occurs the theory of factor analysis is used to realize fault isolation by the matching between the sequences of contribution and carrying degree of score vectors. Finally real vehicle tests verify the effectiveness and realtimeness of the technique.