基于本体的汽轮发电机组故障诊断知识建模

针对目前汽轮发电机组故障诊断领域知识术语复杂、系统异构、知识表示不完备以及共享和重复使用困难等问题,依据故障诊断需求,采用基于本体的知识表示方法,提出了一种适用于汽轮发电机组故障诊断领域的本体构建方法和知识表示模型.在解析了汽轮发电机组故障知识特性的前提下,定义了其本体概念、属性、关系、实例和公理,为知识表示提供了明确的形式化规格说明,并借助Protégé_4.3构建了包含汽轮发电机组的故障类型、故障特征、故障原因和维修策略等故障诊断领域本体,设计了一致性检验的算法.在此基础上,在SQI机械故障综合模拟实验台上模拟汽轮发电机组故障,通过FaCT++推理机实现本体知识推理测试.结果表明基于本体的汽轮发电机组故障诊断知识模型是可行的.

Knowledge modeling of fault diagnosis for turbine generator sets based on ontology

Since the knowledge representation method of fault diagnosis in turbine generator sets is lack of complete in the field of terms, complex terminology, system heterogeneity and difficulty in sharing and reuse etc, a new method for ontology-based knowledge representation is well adopted in terms of fault diagnosis requirements. Methods for ontology construction and a knowledge representation model for the fault diagnosis of turbine generator sets are proposed respectively. The model defines its ontology concepts, attributes, relationships, examples and axioms, and provides a clear formal specification for knowledge representation. Furthermore, Protégé_4.3 is used to construct the ontology of turbine generator sets fault diagnosis domain with fault type, fault characteristic, fault reason and maintenance strategy, and the knowledge is proved to be consistent in the ontology by designing algorithms. Turbine generator sets fault is simulated by SQI mechanical fault comprehensive simulation test bench, and the reasoning test of ontology knowledge is verified by the inference engine named FaCT++. The result indicates that the knowledge model of ontology-based fault diagnosis is feasible.