基于模型诊断的方法，是根据系统模型，来诊断出故障部件。然而，在很多情况下系统模型是未知的，此时需要通过试验，得到系统的部分模型。执行这些试验需要花费大量的资源和时间，因此针对如何缩减试验的次数，利用基于模型诊断求解的思想提出了几个算法来进行部分模型的识别，实验结果验证了本文方法的有效性。-Model-based diagnostic approach is based on system model, to diagnose the failed component. However, in many cases the system model is unknown at this time through testing, to be part of the system model. Perform these tests need to spend a lot of resources and time, so how to reduce the number of trials for the use of model-based diagnosis solving thinking of a few parts of the model algorithm to the identification, experimental results verify the validity of this method.