基于层次广义随机Petri网的测试性建模新方法

针对目前装备系统采用层次化、模块化设计，维修级别与测试性建模复杂度大大提高的问题，提出一种基于层次广义随机Petri网（HGSPN）的测试性建模方法。将主流模型和广义随机Petri网（GSPN）模型进行对比，阐明主流模型存在的问题，以及选择GSPN模型的原因；对装备进行层次划分，建立分层GSPN模型；系统及其组成元件存在多个故障模式，为区分这些故障模式提出一套完整编码方案；给出可达性算法获取层次相关性矩阵，运用测试性评估数学模型得到各层级的测试性水平，将各层级的测试性信息汇总，得到装备完整的测试性水平。以某型导弹发动机系统为例，建立其HGSPN模型，并对测试性指标进行确定，得到100%的故障检测率和66.7%的故障隔离率，验证了所提建模方法和相应算法的有效性。

A New Testability Modeling Method Based on Hierarchical Generalized Stochastic Petri Nets

Since the current equipment system adopts the hierarchical and modular design, the maintenance level and testability modeling complexity are greatly improved. A testability modeling method based on hierarchical generalized stochastic Petri nets (HGSPN) is proposed. The problems existing in the mainstream modeling method and the reasons for selecting the GSPN modeling are analyzed by comparing the mainstream modeling method with the generalized stochastic Petri net (GSPN) modeling method. The equipment is hierarchically divided, and a HGSPN model is established. The multiple failure modes exist in the system and its components. A complete coding scheme is proposed to distinguish these failure modes. The accessibility algorithm is used to obtain the hierarchical dependency matrix, and the testability evaluation mathematical model is used to obtain the testability level. The testability information of each level is aggregated to obtain the testability level of the complete equipment. Taking a missile engine system as an example, the testability indicators are evaluated, and 100% fault detection rate and 66.7% isolation rate are obtained. The effectivenesses of the proposed modeling method and corresponding algorithm are verified.Key