针对采集样本中某些特征缺失无法为已有密闭鼓风炉的故障诊断模型所用的问题,建立了基于案例推理技术的相似度测量和支持向量机的不完备数据故障诊断.通过测量未知特征向量与历史数据库中的样本之间的相似度,检索出最优相似的样本,填补缺失特征值,得到完备的样本.结果表明:该系统能够解决采集样本中特征缺失的问题,具有较高的故障诊断率. Aiming at the problem that some feature missing in the collected samples can not be used in the fault diagnosis model of the closed blast furnace, an incomplete data fault diagnosis based on case-based reasoning technology and support vector machine is established.By measuring unknown feature vector and history The similarity of the samples in the database is retrieved, the best similar samples are retrieved, the missing eigenvalues ​​are filled in, and a complete sample is obtained.The results show that the system can solve the problem of missing features in the collected samples and has a high fault diagnosis rate .