Fault diagnosis of automobile systems using fault tree based on digraph modeling

Abstract

Fault diagnosis of automobile systems is critical, as it adds-up to repair and maintenance time. It is, therefore, desired to make it efficient and effective. One of the conventional approaches is to use the fault tree diagram. But this approach is inadequate with its implicit system structure. Structure of the system means system elements and their interrelations. To alleviate this limitation, a new approach is suggested wherein the structure is in-built, i.e. incorporated explicitly, through digraph modeling that employs a systems approach of graph theory. A system digraph is developed, considering relationships among input and output parameters of subsystems/components of the automobile system in normal and failed conditions. Fault tree of a failure symptom that represents abnormality or a breakdown of the automobile system is obtained from the system digraph. The novelty is extension of the structural approach to automobile systems using digraph model, which has been successfully applied to chemical and process systems. Step-by-step methodology of the structural approach is presented. Its two main two steps are Steps 1 and 2, i.e. ‘Development of Fault tree diagram’ and ‘Diagnosis of fault using the tree diagram’, respectively. The suggested approach is illustrated for hydraulic power steering, an automobile system that is fitted on all current automobiles and particularly, in special purpose vehicles like heavy-duty trucks, earthmovers, dumpers, etc. The suggested approach guides how to diagnose root causes of a fault. The approach is not only helpful to maintenance personnel in effective diagnosis but also in guiding designers in development of reliable automobile systems, accident investigations of automobiles, etc.