The SDG Model Of Fault Diagnosis During AMT Gear Shifting Process

Automated manual transmission （AMT）is a kind of automatic transmission which isbased on dry clutch and mechanical transmission. In recent years, it is widely used in theheavy-duty off-road vehicles for its low cost, high transmission efficiency, reducing drivers’fatigue and many other advantages. AMT is a complex product integration with machinery,electron, fluid and engineering application of automatic theory. Advanced fault diagnosisalgorithm is one of the key technologies of AMT.Signed directed graph（SDG）is a effective method that is used to describe thelarge-scale complicated systems. The research of SDG model becomes hot and achievesmany results in fault diagnosis and safety evaluation in recent years. SDG model shows themutual influencing relationships of different variables clearly by using nodes and directededges, providing model basis for study of fault propagation as well as finding the faultsource timely. This paper quotes SDG model to fault diagnosis of AMT gear shiftingprocess. The specific is as follows:Firstly, function and common classification of automatic transmission are introduced,and also the future trend. Then the development of vehicle and AMT fault diagnosistechnology is described in detail.Secondly, AMT is classified by different driving type. The structure and principle ofresearch platform—electronically control and hydrodynamic drive AMT is especiallyanalyzed here. AMT gear shifting process is divided into six subphases based on logicalsequence，clutch separation process, main gearbox neutral gear shifting process, auxiliarygearbox gear shifting process, main gearbox gear selecting process, main gearbox gearshifting process and clutch engagement process, working process of each subphase isanalyzed in detail. Common faults of AMT gear shifting process is described by differentsubphases.Then mathematical models for the six subphases are built from aspects of vehicletransmission system dynamics and automatic control system kinematics. SDG healthymodels for the six subphases are also established based on mathematical models andrelevant experience. SDG healthy models are simplified for the limit of measurementmethod or economic condition and also based on relevant simplified method.Fault examples of AMT gear shifting process are diagnosed by using qualitative andsemi-quantitative SDG model method. At last, Simulink model of fault diagnosis for AMT gear shifting process is established based on SDG simplified model, thus intelligentdiagnosis by using computer is realized. The result shows that it is feasible and effective touse SDG model for gear shifting in AMT.