摘要
针对传统模糊Petri网在对不确定环境下的专家系统的知识表示与推理时无法兼顾不确定知识的模糊性与随机性、在复杂的故障情况下故障的因果关系表达不清晰、定量推理计算时缺乏层次性、不能局部求解的问题,构建一种基于云模型的分层模糊Petri网以加强模糊Petri网的知识表示能力和提高推理过程的计算效率。利用专家知识和Petri网层次分解原则将系统故障模式和故障原因之间的因果关系进行建模,使故障建模更具结构性,计算更加灵活;应用云模型处理知识的模糊性和不确定性;通过合理考虑局部权重和全局权重,结合Petri网层次分解原则和云聚合算子给出相应的推理算法。实例验证表明,所提方法能够有效对系统进行风险评估,且在知识表示和推理方面优于其他方法。
Aiming at the problems that traditional fuzzyPetri nets can't give consideration to the fuzziness and randomness of uncertain knowledge,the expression of causality in complex fault,quantitative reasoning is lack of hierarchy and can not be solved locally,a hierarchical fuzzy Petri net based on cloud model is constructed to strengthen fuzzy Petri net The ability of knowledge representation and the efficiency of reasoning process are improved.The causal relationship between system failure modes and failure causes is modeled by expert knowledge and Petri net hierarchical decomposition principles,which makes the fault modeling more structured and more flexible;the cloud model is applied to deal with the fuzziness and uncertainty of knowledges;the corresponding reasoning is given by reasonably considering local weight and global weight,combining with Petri net hierarchical decomposition principles and cloud aggregation operators.Example verification shows that the proposed method can effectively assess the risk of the system,and is superior to other methods in knowledge representation and reasoning.
作者
古莹奎
何力韬
毕庆鹏
GU Ying-kui;HE Li-tao;BI Qing-peng(School of Mechanical and Electrical Engineering,Jiangxi University of Science and Technology,Jiangxi Ganzhou 341000,China)
出处
《机械设计与制造》
北大核心
2024年第2期369-372,379,共5页
Machinery Design & Manufacture
基金
国家自然科学基金资助项目(61963018)
江西省自然科学基金资助项目(20181BAB202020)。
