论文部分内容阅读
汽轮机故障诊断的研究对于提高汽轮机运行的安全性，经济性及可靠性具有重要意义，本文基于事例推理提出了研究凝汽设备故障的新方法。总结了基于事例推理的优点，研究了凝汽设备的工作过程，分析了凝汽设备常见故障机理。给出了基于事例推理过程的流程图，详细介绍了基于事例推理进行故障诊断的实现过程。在建立知识库时引入了置信度；基于反向传播神经网络进行事例检索；给出了诊断结果的证实规则和解释方法。最后，将该方法用于某电厂３００ＭＷ汽轮机组凝汽系统故障诊断的研究，取得了满意的诊断结果。总之，该方法尤其适合于存在大量事例却又难于用模型或算法描述的应用领域。
The research of turbine fault diagnosis is of great significance to improve the safety, economy and reliability of turbine operation. Based on case-based reasoning, this paper presents a new method to study the fault of condensing equipment. The advantages of case-based reasoning are summarized, the working process of condensing equipment is studied, and the common failure mechanism of condensing equipment is analyzed. The flow chart based on case-based reasoning is given, and the realization process of fault diagnosis based on case-based reasoning is introduced in detail. In the establishment of knowledge base, the reliability was introduced. Case retrieval was carried out based on backpropagation neural network. The rules and explanation of diagnosis were given. Finally, the method is applied to the fault diagnosis of condensate system of a 300MW steam turbine unit in a power plant, and satisfactory results are obtained. In summary, this method is particularly suitable for applications that have a large number of instances but are difficult to describe using models or algorithms.