《Table 3 Actual fault data classification results》
提示：宽带有限、当前游客访问压缩模式
本系列图表出处文件名：随高清版一同展现
《Research on fault diagnosis method of piston rod based on harmonic wavelet and manifold learning》
In order to verify the effectiveness of the proposed method in the fault diagnosis of piston rod，BP neural network is used to identify the fault state.For the four conditions，100 sets of sample data are selected as training samples and 200 sets of data are used as test samples respectively.After analysis，the number of hidden layers is 8.The sensitive features obtained by harmonic wavelet+LTSA dimension reduction，harmonic wavelet+PCA dimension reduction and PCA dimension reduction are input and trained by BP neural network.The results are shown in Table 3.
| 图表编号 | XD0020601300 严禁用于非法目的 |
|---|---|
| 绘制时间 | 2018.09.01 |
| 作者 | 江志农、Zhu Lina、Zhang Jinjie、Zhou Chao |
| 绘制单位 | Diagnosis and Self-recovering Engineering Research Center,Beijing University of Chemical Technology、Diagnosis and Self-recovering Engineering Research Center,Beijing University of Chemical Technology、Compressor Health and Intelligent Monitoring Center of |
| 更多格式 | 高清、无水印（增值服务） |
