摘要
针对车辆发动机故障诊断的问题,首先通过希尔伯特-黄变换(HHT)分析处理发动机振动信号的时频特征,提取出基于边际谱能量变化的故障特征参数来准确的表征发动机故障现象。然后利用BP神经网络对发动机故障数据特征进行训练,并对多种实测的故障数据进行测试和识别,验证所提出的算法的准确性。
Aiming at the vehicle engine fault diagnosis issue, applies Hilbert Huang transform （HHT） on engine vibration signal for time-frequency characteristics, and extracts parameters based on change of marginal spectrum energy to accurately characterize the engine fault phenome-non. Then, the BP neural network is used to train the engine fault data, and the accuracy of the proposed algorithm is verified by test data.
关键词
故障诊断
发动机
HHT
BP神经网络
Fault Diagnosis
Engine
Hilbert-Huang Transform
BP Neural Network
