论文部分内容阅读
柴油机振动信号具有非平稳性,用最优小波包将不同故障的振动信号分解到不同频段。提取各频段的能量组成特征向量输入SOM-BP神经网络,通过神经网络输出结果判别柴油机的故障类型。与BP网络的训练结果相比较,证明将最优小波包分解与SOM-BP神经网络相结合的方法可以得到更好的分类结果,有一定的工程实用性。
Diesel engine vibration signal has non-stationary, the best wavelet packet will be different failure of the vibration signal decomposition to different frequency bands. The energy components of each frequency band are extracted and input into the SOM-BP neural network. The output of the neural network is used to determine the type of diesel engine fault. Compared with the training results of BP network, it proves that the method of combining the optimal wavelet packet decomposition and SOM-BP neural network can get better classification results and has certain engineering practicability.