Research On System-level Testability Allocation Method For Equipment

With the increasing of complexity and functions of weapon equipments, challenges are brought to system test and maintenance. Design for Testability (DFT) is one of key technologies that enhance test and diagnosis capability of systems. Testability allocation is an important process in DFT. It allocates given system-level testability indices to lower levels, which determines testability indices of sub-systems, so that requirements for sub-systems in DFT are specified. Therefore research on testability allocation is important for DFT.The existing testability allocation methods (TAM) are short of theoretical analysis, and their implementation processes have some demerits such as subjectivity, the allocated results needing manual adjustment. To solve these problems, this dissertation researches on theory and method of testability allocation. The main contents and results are as follows:(1) Aiming at the problem that the applied range of functions of testability indices is restricted, testability block diagram is introduced. Testability block diagrams in the form of series, parallel and series-to-parallel are established for variant systems. The corresponding functions of testability indices are analyzed. Then aiming at the problem that the allocation model is improper, allocation models of fault detection rate (FDR) and fault isolation rate (FIR) are put forward. Then the allocation problem is reduced to calculate allocating weights in the model.(2) To calculate allocating weights, impact of representative factors which influence testability allocation is analyzed. These factors include failure rate, significance degree, complexity degree, mean time to repair (MTTR) and the difficulty degree of fault detection or isolation. Then these factors are quantified. The quantified results act as input data of allocation.(3) The TAM based on Analytic Hierarchy Process (AHP) is researched. To solve the demerits of traditional AHP, generation approach of judgment matrix is improved, so that AHP can be applied to testability allocation. Then the synthetical importance degree is calculated by AHP, and then allocating weights in restricted range are calculated based on synthetical importance degree. Thus allocation of testability indices is implemented. The validity of this testability allocation method is validated by a case. Compared to other allocation methods, the allocating results of this method is better.(4) A software tool for testability allocation is compiled. As an application, testability allocation on a mechatronic track and stable servo platform is carried out. The results show that the TAM based on AHP and the software tool are effective.