论文部分内容阅读
针对系统结构复杂、维修保障难度大,多具有不确定性(不完整性)信息、多故障和关联故障等特点,融合贝叶斯理论和策略优化算法的优势,提出一种混合快速诊断算法.将诊断任务分解成若干单故障子任务,对子任务用故障树和贝叶斯网络分别建模,根据贝叶斯-蒙特卡洛算法获得先验概率,进行诊断推理分析,获得系统故障薄弱环节,同时运用综合策略优化算法,对故障原因进行诊断策略优化,以实现快速诊断为目的.以电控发动机复杂系统为例,对混合快速诊断算法进行应用研究,验证了该算法的正确性和有效性.
In view of the complex system structure, the difficulty of maintenance and support, the information of multiple uncertainties (incompleteness), the multiple faults and the associated faults, this paper proposes a hybrid rapid diagnostic algorithm by combining the advantages of Bayesian theory and strategy optimization algorithm. The diagnostic task is decomposed into a number of single fault subtasks. The subtasks are respectively modeled by a fault tree and a Bayesian network. The prior probabilities are obtained according to the Bayesian - Monte Carlo algorithm, and the diagnostic reasoning analysis is performed to obtain the fault of the system fault , And at the same time use the comprehensive strategy optimization algorithm to optimize the diagnosis strategy for the cause of the fault so as to achieve the purpose of rapid diagnosis.The application of the hybrid rapid diagnosis algorithm to the complex system of the electronically controlled engine is applied to verify the correctness and effectiveness of the algorithm Sex.