摘要
结合信息融合的一般模型和层次结构,构建了电动机故障的信息融合诊断模型。该模型涵盖了数据融合的数据层、特征层和决策层三个层次,充分利用电动机三相电流、电压和绕组温度等多种故障信息,基于Park矢量,进行了多层次的融合;采用了基于人工神经网络、小波分析和证据理论等的两级诊断策略,以进一步提高转子断条、气隙偏心、绕组匝间短路等电动机常见故障的诊断准确性和灵敏性。
According to the general model and architecture of information fusion, a new model is presented to diagnose asynchronous motor＇s faults. The model consists of data fusion level, characteristic fusion level and decision-making fusion level, its reasoning strategy is divided into two steps by integrating Park Vector method, ANN, WT and D - S Evidence Theory. Based on the model, motor＇ s faults such as rotor broken-bar, gap eccentricity and inter-turn short circuit will be more accurately and sensitively diagnosed.
出处
《电气应用》
北大核心
2006年第9期63-66,共4页
Electrotechnical Application
基金
国家自然科学基金
项目号50504015
中国矿业大学科研基金
项目号0C4499
关键词
信息融合
故障诊断
异步电动机
证据理论
information fusion fault diagnosis asynchronousmotor evidence theory
