论文部分内容阅读
故障诊断专家系统的推理策略取决于诊断知识的表达方式。基于模糊Petri网的知识表达是库所、命题、变迁、输入、输出及其可信度、影射等的集合。基于模糊产生式规则的模糊Petri网模型,能较好地表示和处理模糊知识,并能根据变迁的可信度大小,采用深度优先的搜索方法,以反向并行推理来缩小搜索空间,寻找到一条最有可能产生已知故障的路径。为此,以机车柴油机故障信息诊断系统为例,针对其故障信息的不确定性和模糊性,给出了基于Petri网的模糊诊断知识表达方法,讨论了建立在模糊Petri网模型基础上的反向模糊推理算法。
The reasoning strategy of a fault diagnosis expert system depends on how the diagnostic knowledge is expressed. Knowledge representation based on fuzzy Petri nets is a collection of places, propositions, changes, inputs and outputs, their credibility, reflection and so on. Fuzzy Petri net model based on fuzzy production rules can express and deal with fuzzy knowledge well, and can reduce the search space according to the credible degree of change, depth-first search method and reverse parallel reasoning A path most likely to cause a known failure. Therefore, taking locomotive diesel fault information diagnosis system as an example, aiming at the uncertainty and fuzziness of its fault information, a fuzzy diagnosis knowledge representation method based on Petri nets is given, and the countermeasures based on fuzzy Petri net model are discussed To fuzzy reasoning algorithm.