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：
为提取走行部故障信号中的冲击成分,提出基于集合经验模态分解降噪和流形学习的故障特征提取模型。依据由峭度和互相关系数所得的综合指标最大准则对故障信号进行集合经验模态分解降噪,以突出各故障主要冲击特征信息;提取已降噪信号的时域、频域、小波包能量矩等多个特征来构造每个样本的高维特征集;运用邻域保持嵌入算法进行维数约简;利用支持向量机进行故障类型识别。标准数据集和高铁故障数据仿真实验结果验证了该模型的有效性。
英文摘要：
To extract the impact components caused by running gear fault signals,based on ensemble empirical mode decomposition（EEMD）denoising and manifold learning,a new feature extraction model was proposed.Firstly,the fault signals were denoised by EEMD to highlight the main impact characteristics,based on the maximal comprehensive index obtained by kurtosis and correlation coefficient.Secondly,multiple features of denoised signals such as time domain,frequency domain,wavelet packet energy moments were extracted to form high dimensional feature set of each sample.Thirdly,the neighbor preserving embedding（NPE）algorithm was employed for dimensionality reduction.Finally,the support vector machine（SVM）was used for recognizing fault types.The validity of the proposed method was verified by the standard data set and the experimental results of high-speed train fault data simulation.
同期刊论文项目
同项目期刊论文
期刊信息
位置：