Diagnostics Structural Algorithms using RODON

Time to market for new products decreases ever more due to the global competitive pressure. A company’s market share depends largely on its capability to satisfy the ever increasing customer requirements with respect to functionality and reliability. The shortened development times and the increasing complexity of the products - as indicated by the significantly increasing electrical and electronic content - may lead to difficulties if not handled appropriately.

Despite careful development and construction, some of these components may eventually fail. To avoid unnecessary damage, for example environmental or financial, there is a need to locate and diagnose these faults as fast as possible. This can be done with a diagnostic system, which should produce an alarm if there is a fault in the mechanical system and, if possible, indicate the reason behind it. The objective of the project is to investigate possibilities to utilize modern theory and methodology, from the research field model based diagnosis, in connection with the commercial tool RODON. RODON provides systematic, scalable and industry proven diagnostic methods for both vehicle on-board diagnostics as well as manufacturing and off-board service bay diagnostics. RODON computes diagnostics results by simulating the system in various operating modes combined with failure modes. Hereby it generates explicit information on the interplay of
the product’s components and the system behavior. Especially, RODON can systematically compute symptoms or top events from root causes of a malfunction and vice versa, i.e. finds the root cause of a failure from symptoms (events). It derives reliability analyses and diagnostic knowledge from these root cause of a failure symptom (event)relationships.

The main objective of this thesis proposal is to provide a completely automated construction process of model based diagnostics systems that employs structural algorithms. From a given RODON model the underlying equations systems that describes the behavior of the mode need to be extracted. The structure of the diagnosis model is then used to find a set of testable models for deriving a sound and complete diagnostic system. The algorithms deriving a sound diagnostic system have been previously developed at the Division of Vehicular System, at Linköping University. Despite the fact that these algorithms have been developed with the aim of automating the model based diagnosis process many industrial users are still deriving the diagnostic system manually.
This is an error-prone and time consuming task. Therefore it is imperative to provide tool support to automate this process.

The master student working on this project should identify, review, and evaluate structural algorithms for diagnostic system design previously developed at the Division of Vehicular System, Linköping University, the in the context of RODON simulation models. The following key questions must be addressed:
• What structural algorithms are most suitable to RODON based diagnostics models?
• How this methods should be integrated in an existing diagnostics environment?
• How the results of the structural model diagnostics should be interpreted and presented to the user?

At the end of the project it is expected that a prototype plug-in implementation is successfully developed and integrated into RODON as a proof of concept.

In order to test and verify the efficiency of the proposed structural based algorithms in the context of model based diagnosis, it is highly desirable that an empirical validation on real world industrial models is performed at the end of the project. The test model will be provided
by the Industrial partner of this project.

It is the aim of the Master thesis to demonstrate that the student is capable of coming to terms with an industrial project task within a reasonable period of time and to handle it in a scientifically sound manner. At the end of the project, it is compulsory for