Research On Key Technologies For Estimating Softwre In Reliability Analysis Of Vehicle

With the rapid economic development, the number of cars and other vehicles rises sharply, however, the rapid expansion of car consumption and production capacity bring enormous challenges to the quality of car. So the vehicle reliability assessment software platform has been concerned by domestic and foreign researchers. This system is the integration of the two technologies-the Analysis technology based on truncated BDD and the dynamic gate simulation technology. It not only offers the quantitative analysis of the vehicle by calculating the probability of the various failure modes, but also plays an important role in improving the efficiency of vehicle fault diagnosis and reliability design.To meet the needs of the assessment, this thesis is designed a three-layer framework of vehicle estimating system, which includes such three parts as failure database, middleware of reliability analysis and graphical modeling tool.Specifically, this thesis includes the following aspects:1. To meet the functional requirements of the system, this thesis is established the overall framework of the software. It can be divided into three main units, such as database layer, Algorithm layer and the user interaction layer. The user interaction layer is mainly in the charge of the fault tree modeling and the feedback of the analysis results. The algorithm layer is the core module of the software, which make itself to been chose by the user in the form of algorithm library. The database layer is responsible for the realization of the template library and the failure database. On this basis, the ideas of modular and hierarchical are used during the server software and its sub-unit design.2. Choosing the automobile as the example, we study the failure mode analysis method in this thesis, based on the structure of the vehicle. This thesis is briefly introduced the vehicle structure and core subsystems which include the engine, chassis, electrical control and battery system. Then, we do analyze the failure modes of these subsystems from the structure and function and build the corresponding fault trees. Due to space limitations, in this thesis, we just build fault trees for parts of the electronic control system and braking system. And we select these trees as the test samples.3. Fault tree approximate analysis based on the truncated BDD is the research focus of the article. Truncated BDD is a current research focus of the approximate analysis for large-scale static fault tree. The thesis is pointed out that there is a many-to-one performance issue between the BDD operation and sub-BDD in the existing truncated BDD algorithm. And then, we design to improve the truncated BDD algorithm based on the write-back mechanism. The experiment data show that the improved algorithm can effectively enhance the BDD analysis performance. Based on the improvement of truncated BDD algorithm, we use iterative approximation strategy to design reliability interval approximation of a static fault tree model of vehicle reliability analysis methods combining with the zero truncated and one truncated.4. We does the preliminary study on the dynamic fault tree based on simulation technique in this thesis. It discusses the four basic dynamic gates, such as PAND gate, SEQ gate, SPARE gate, FDEP gate. Since modern vehicles have spare-fast repair mechanism, we introduce a new gate, redundancy repairable gate (RRG (n, m, k)), and analyze the dynamic reliability behavior of the dynamic gate, and finish the simulation process. In the end, we use the simulation technology to analyze the braking system of the car.