风电机组故障知识的获取表达与推理框架

Acquisition,expression and reasoning framework of wind turbine fault knowledge

针对目前风电机组故障种类多、故障知识关联关系复杂、知识表达差异化大、知识推理效率低等问题,提出了风电机组故障知识的获取表达与推理框架。首先,通过基于故障树分析方法的故障模式及影响分析方法,全面获取并梳理出风电机组故障排查与检修维护的专家知识;然后,借助本体理论将非结构化的专家知识进行结构化表达,形成知识图谱,将知识可视化展示;同时,结合本体自定义规则以及因果推理模型实现故障原因的查询推理,提高了知识查询与推理的效率;最后,通过具体的机组故障案例说明了本方法的实用性。研究结果可为风电场运维检修工作的智能化发展提供方向。

Aiming at solving the problems of large number of wind turbine faults, complex fault knowledge relationship, large difference of knowledge expression and low efficiency of knowledge reasoning, a framework of acquisition, expression and reasoning of wind turbine fault knowledge is proposed. Firstly, through the failure mode and effect analysis method based on the fault tree analysis method, the expert knowledge of wind turbine trouble shooting and maintenance is comprehensively obtained and sorted out. Then, with the help of ontology theory,unstructured expert knowledge is expressed structurally to form a knowledge map and displayed visually. Combined with self-defined rules of ontology and causal reasoning model, the query and reasoning of fault causes are realized,which improves the efficiency of knowledge query and reasoning. Finally, the practicability of this method is illustrated by a specific unit fault case. The results of this study can provide a direction for the intelligent development of wind farm’s operation and maintenance.