欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对齿轮泵故障成因复杂、模糊性强的特点,结合小波包分解与K-L变换,提出一种适用于支持向量机故障诊断的特征提取方法。通过小波包对样本故障振动信号进行分解得到特征向量,而后利用K-L变换处理得到新的特征向量集,达到降维去噪的目的。将处理后的特征向量集用于支持向量机的模型训练,分析结果表明：该方法能够有效提高故障模式识别准确率和识别效率。
英文摘要：
According to the characteristics of the gear pumps fault of complicated formation, strong fuzziness, combined with wavelet packet decomposition and K-L transform, presented a suitable support vector machine fault diagnosis feature extraction method The feature vector was decomposed by wavelet packet, and the new feature vector set was processed by K-L transform. The feature vector set was used to support vector machine model training. The result shows that the method can effectively improve the accuracy and efficiency of fault pattern recognition.
同期刊论文项目
同项目期刊论文
期刊信息
位置：