Graphical Hybrid Diagnostic Model-Based Software Architecture for System Testability Analysis

With the complexity and integration of equipment system structure, the difficulty of fault discovery and location is greatly increased, so it is important to consider system testability in equipment design and research and development. There are mainly two kinds of models used in system testability modeling: multi-signal flow model and mixed diagnosis model. The mixed diagnosis model can make up for the lack of multi-signal flow model to reflect the relationship between the fault and the function of each module of the system, and can meet the needs of testability modeling more comprehensively. Based on the graphical hybrid diagnosis model, this paper proposes an implementation architecture of testability analysis software, which realizes the establishment of the graphical hybrid diagnosis model, the generation of the failure-test matrix, the calculation of testability index, and the optimization based on the classical weight function and the Rollout algorithm to obtain the fault isolation sequence. It is verified on two concrete case models that the testability index obtained by our software is consistent with the advanced foreign software eXpress, which proves the rationality of the software architecture.