油料装备故障智能诊断平台研究

为了解决油料装备故障智能诊断问题,构建由平台支持层、技术支持层、数据层和应用层组成的故障智能诊断平台.应用模糊神经网络、专家系统、微粒群优化算法和证据理论作为智能融合工具,对检测数据实现数据级、特征级和决策级三级数据融合,借助数据间互补信息,降低检测数据误差,消除冗余信息.建立基于案例推理CBR和规则推理RBR的故障诊断集成推理机.将最近邻方法与灰色关联理论相结合,提出分层检索匹配算法.

The PlAtform of Intelligent FAult DiAgnosis for Oil Equipment

In order to solve the problems of fault diagnosis, the platform of intelligent fault diagnosis is built, which is divided into four levels, such as platform and technical support level, data level, application level. The detected data is fused using Fuzzy Neural Network, Expert System, Particle Swarm Optimization and Evidence Theory, so that three -phase data fusion （i. e. data, feature and decision -making） is implemented. By using complementary information between the data, the data errors are reduced, and the redundant information is eliminated. The inte- grated inference engine is designed via Case -based Reasoning and Rule -based Reasoning. By combing the nea- rest neighbor method with the gray relational theory, the match algorithm based on the hierarchical retrieval is pro- posed.