基于Petri网模糊推理算法,引入改进粒子群优化算法,使其具有参数学习能力,摆脱系统中众多未知参数对专家和经验的依赖.以舰船航空保障系统为研究对象,建立了基于具有学习能力FPN(Fuzzy Petri net)的系统易损性评估模型,算例佐证了其学习效率及精度.与传统系统易损性评估方法的对比试验表明该评估模型的全面性和合理性,并可推广应用于复杂舰船系统易损性评估研究. Based on the Petri nets fuzzy inference algorithm, the improved Particle Swarm Optimization (PSO) algorithm is introduced to make it have the ability of parameter learning and get rid of the dependence of many unknown parameters on the experts and experience in the system.With the research of ship aviation security system, FPN (Fuzzy Petri net) system vulnerability assessment model, an example to demonstrate its learning efficiency and accuracy.Compared with the traditional system vulnerability assessment methods show that the evaluation model is comprehensive and reasonable, and can be extended to use Research on Vulnerability Assessment of Complex Ship System.