针对流形学习算法的增量处理问题，提出一种邻域自适应增量式PCA-LPP流形学习算法，阐述了算法的基本原理以及增量样本处理方法。对新增样本的引入，首先根据已有样本对协方差矩阵和相似矩阵进行增量更新，而后结合已有样本降维结果对新增样本降维结果进行估计，最后采用子空间迭代法实现新旧样本降维结果的更新。采用齿轮箱故障信号特征向量对邻域自适应增量式PCA-LPP流形学习算法进行检验，结果表明，邻域自适应增量式PCA-LPP流形学习算法降维后特征具有良好的故障分类识别效果。
Abstract
In view of the incremental learning problem of manifold learning algorithm, the adaptive neighborhood incremental PCA-LPP manifold learning algorithm is presented, and incremental learning principle of algorithm is introduced. For incremental sample data, The adjacency and covariance matrix was incremental updated by existing sample, then dimensionality reduction results of incremental sample was estimated by of dimensionality reduction results existing sample and updated matrix, finally, the dimensionality reduction results of incremental and existing sample was updated by subspace iteration method. The adaptive neighborhood incremental PCA-LPP manifold learning algorithm was application in processing of gearbox fault signals, the dimensionality reduction results by incremental learning have very small error compared with batch learning, spatial aggregation of incremental sample was basically stable, and fault identification rate was increased.
关键词
增量式学习 /
自适应 /
流形学习 /
故障诊断
{{custom_keyword}} /
Key words
incremental learning /
adaptive /
manifold learning /
fault diagnosis
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 赵振华，郝晓弘. 局部保持鉴别投影及其在人脸识别中的应用[J]. 电子与信息学报. 2013.35(2):463-466.
ZHAO Zhenhua, HAO Xiaohong. Linear Locality Preserving and Discriminating Projection for Face Recognition[J]. Journal of Electronics & Information Technology. 2013.35(2):463-466.
[2] 肖婷，汤宝平，秦毅，陈昌. 基于流形学习和最小二乘支持向量机的滚动轴承退化趋势预测[J]. 振动与冲击. 2015.34(9):149-153.
XIAO Ting, TANG Bao-ping, QIN Yi, CHEN Chang. Degradation trend prediction of rolling bearing based on manifold learning and least squares support vector machine[J]. Journal of Vibration and Shock. 2015. 34(9):149-153.
[3] X F He, P Niyogi. Locality preserving projections[A]. Neural Information Processing Systems16[C]. Vancouver. MIT Press, 2004.153-160.
[4] Zhenhua Zhang, Xinzhong Zhu, et al. Image Retrieval Based on PCA-LPP[A]. 2011 10th International Symposium on Distributed Computing and Applications to Business, Engineering and Science[C]. Wuxi, Jiangsu, China. 2011: 230-233
[5] 王健，冯健，韩志艳. 基于流形学习的局部保持PCA算法在故障检测中的应用[J]. 控制与决策. 2013.28(5):684.
Wang Jian, Feng Jian, Han Zhiyan. Locally preserving PCA method based on manifold learning and its application in fault detection[J]. Control and Decision. 2013.28(5):684.
[6] 张晓涛，唐力伟，王平，邓士杰. 基于多尺度正交PCA-LPP流形学习算法的故障特征增强方法[J]. 振动与冲击. 2015.34(13):66-70+114.
Zhang Xiao-tao, Tang Li-wei, Wang Ping, Deng Shi-jie. Fault feature enhancement method based on multiscale orthogonal PCA- LPP manifold learning algorithm[J]. Journal of Vibration and Shock. 2015.34(13):66-70+114.
[7] 杨庆，陈桂明，童兴民，何庆飞. 增量式局部切空间排列算法在滚动轴承故障诊断中的应用[J]. 机械工程学报. 2012.48(5): 81-86.
YANG Qing, CHEN Guiming, TONG Xingmin, HE Qingfei. Application of Incremental Local Tangent Space Alignment Algorithm to Rolling Bearings Fault Diagnosis[J]. Journal of Mechanical Eigineering. 2012.48(5):81-86.
[8] Peng Jia, Junsong Yin, Xinsheng Huang, Dewen Hu. Incremental Laplacian eigenmaps by preserving adjacent information between data points[J]. Pattern Recognition Letters. 2009.30(8):1457-1463.
[9] CHIN T J, SUTER D. Out-of-sample extrapolation of learned manifolds[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.30(9):1547-1556.
[10] 谈超，关佶红，周水庚. 基于等角映射的多样本增量流形学习算法[J]. 模式识别与人工智能. 2014.27(2):127-133.
TAN Chao, GUAN Jihong, ZHOU Shuigeng. Multi-sample Incremental Manifold Learning Algorithm Based on Isogonal Mapping[J].PR & AI. 2014.27(2):127-133.
[11] 孙即祥. 现代模式识别[M]. 长沙：国防科技大学出版社，2002.
[12] Shuicheng Yan, Dong Xu, Benyu Zhang, et al. Graph Embedding and Extensions - A General Framework for Dimensionality Reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2007.29(1):43.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(788 KB)
