基于贝叶斯网络的电容型设备故障诊断方法

Fault diagnosis based on Bayesian network for capacitive equipment

提出了基于贝叶斯网络的电容型设备故障诊断方法,阐述了贝叶斯网络的构建过程。通过广泛收集有关电容型设备的故障资料,综合其各种检测数据和故障征兆,获得了较为全面的故障集和征兆集。经过对数据的统计分析获得了各故障类型下各征兆量有明显体现的条件概率,在此基础上建立了基于贝叶斯网络的电容型设备故障诊断模型,并根据电容型设备故障诊断的特点改进了贝叶斯网络的推理过程,采用连概率计算过程进行故障类型的概率信息计算,根据概率信息进行故障分类,提高了该方法的实用性。通过电容型设备故障实例分析,诊断结果验证了该方法的正确性和有效性。

A fault diagnosis method based on Bayesian network is presented for capacitive equipment. The construction process of Bayesian network is expounded. The information of capacitive equipmenf fault are widely collected, and various detected data and fault symptoms are summarized to form a comprehensive fault set and symptom set. The conditional probability of each symptom clearly presented under different fault type is acquired,based on which, the fault diagnosis model based on Bayesian network is founded for capacitive equipment and the reasoning process of Bayesian network is improved. The probability calculation method is adopted and the fault type is diagnosed by the calculated probability, which improves its practicability. Its correctness and effectiveness are validated by the practical fault diagnosis results for capacitive eauinment.