基于BN和改进DST的轮毂电机故障诊断方法

In-wheel motor fault diagnosis method based on BN and improved DST

为实现不同信号来源下的轮毂电机故障诊断,提出一种基于贝叶斯网络(BN)和改进DST理论的轮毂电机故障诊断方法.以电动汽车轮毂电机运行安全为目标,首先,搭建基于振动和噪声的BN诊断模型,得到不同运行工况下基于振动和噪声的BN诊断模型后验概率;然后,提出基于熵权值的改进DST,对基于振动和噪声的BN诊断模型后验概率的冲突部分进行重新分配,得到新的基本信度函数;最后,通过轮毂电机台架实验来验证该方法的有效性.结果显示:改进DST在解决证据间冲突问题的同时,能够有效融合基于振动和噪声的BN诊断后验概率,实现轮毂电机故障诊断.

In order to realize in-wheel motor fault diagnosis under different signal sources, an in-wheel motor fault diagnosis method based on Bayesian network(BN) and improved DST(Dempster-Shafer evidence theory) was proposed.The BN diagnostic model based on vibration and noise was firstly built for the operation safety of electric vehicle in-wheel motor. Secondly, the posterior probability of BN diagnosis model based on vibration and noise under different operating conditions was fused.Then an improved DST based on entropy weight was proposed to reallocate the conflicting parts of the posterior probability of BN diagnostic model based on vibration and noise, and a new basic reliability function was obtained. Finally, the effectiveness of the method was verified with in-wheel motor bench test.It is shown that improved Dempster-Shafer evidence theory can effectively solve the problem of inter-evidence conflict and integrate the BN diagnosis posterior probability based on vibration and noise to realize inwheel motor fault diagnosis.