基于优化特征的集成变量预测模型及其应用

doi:10.3969/j.issn.1004-132X.2021.18.002

摘要/Abstract

摘要： 针对传统基于变量预测模型的模式识别方法在小样本情况下识别精度较低的问题，提出一种基于递归量化特征优化选择的集成变量预测模型（EVPM）。该模型采用递归量化分析对被分析信号进行特征提取，在交错式最大权值最小冗余规则下选择权重高且冗余度低的优化特征子集，引入高斯函数、径向基函数、广义回归函数建立特征变量间的复杂非线性关系模型，根据各模型的拟合误差进行自适应加权集成。实验结果表明，基于优化特征EVPM的模式识别方法即使是在小样本的情况下也能够有效地进行滚动轴承故障类型与故障程度的自动识别，且在精度和稳定性上明显优于同类型传统方法。

关键词: 特征优化选择, 变量预测模型, 加权集成, 故障诊断

Abstract: Aiming to the problems of low recognition accuracy of conventional variable predictive model resulting in case of small samples, an integrated variable predictive model was proposed based on the optimal selection of recurrence quantification features. Some feature subsets with high weight and low redundancy optimally were formed and then the best feature subsets were selected using the embedded rater. The Gauss function, radial basis function and generalized regression functions were introduced to establish the complex nonlinear interaction relationships among the selected features. The newly established models and conventional models were integrated based on the fitting errors of each model. The experimental results show that the proposed method has higher accuracy and better stability in the bearing fault diagnosis compared with the conventional methods, especially in case of small samples.

Key words: optimal feature selection, variable predictive model, weighted integration, fault diagnosis

中图分类号:

TH113

刘小峰, 谭奇, 叶蓉婷. 基于优化特征的集成变量预测模型及其应用[J]. 中国机械工程, 2021, 32(18): 2153-2158，2164.

LIU Xiaofeng, TAN Qi, YE Rongting. Ensemble Variable Predictive Model Based on Optimal Features and Its Applications[J]. China Mechanical Engineering, 2021, 32(18): 2153-2158，2164.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/10.3969/j.issn.1004-132X.2021.18.002

http://www.cmemo.org.cn/CN/Y2021/V32/I18/2153

参考文献

［1］RAGHURAJ R, LAKSHMINARAYANAN S. Variable Predictive Models—a New Multivariate Classification Approach for Pattern Recognition Applications［J］. Pattern Recognition, 2009,42(1):7-16.
［2］杨宇, 李紫珠, 何知义, 等. QGA-VPMCD智能诊断模型研究［J］. 振动与冲击, 2015, 34(13):31-35.
YANG Yu, LI Zizhu, HE Zhiyi, et al. Research on QGA-VPMCD Intelligent Diagnosis Model ［J］. Vibration and Shock, 2015, 34(13):31-35.
［3］LUO S, CHENG J, ZENG M, et al. An Intelligent Fault Diagnosis Model for Rotating Machinery Based on Multi-scale Higher Order Singular Spectrum Analysis and GA-VPMCD［J］. Measurement, 2016,87:38-50.
［4］刘吉彪, 程军圣, 马利. 基于PSODACCIW-VPMCD的滚动轴承智能检测方法［J］. 振动与冲击, 2015， 34(23):42-47.
LIU Jibiao, CHENG Jusheng, MA Li. Intelligent Detection Method of Rolling Bearing Based on PSODACCIW-VPMCD［J］. Vibration and Impact, 2015, 34(23):42-47.
［5］柏林, 曾柯, 徐冠基, 等. 基于RQA和V-VPMCD的滚动轴承故障识别方法［J］.振动·测试与诊断, 2018, 38(2):314-319.
BO L, ZENG Ke, XU Guanji, et al. Rolling Bearing Fault Identification Method Based on RQA and V-VPMCD ［J］. Vibration·Test and Diagnosis, 2018, 38(2):314-319.
［6］宋坤骏. 基于改进VMD、ELM和VPMCD算法的滚动轴承故障诊断方法研究［D］. 成都：西南交通大学, 2018.
SONG K. Research on Rolling Bearing Fault Diagnosis Method Based on Improved VMD, ELM and VPMCD Algorithm［D］. Chengdu:Southwest Jiaotong University, 2018.
［7］高佳程, 曹雁庆, 朱永利, 等. 基于KELM-VPMCD方法的未知局部放电类型的模式识别［J］. 电力自动化设备, 2018(5):141-147.
GAO Jiacheng, CAO Yanqing, ZHU Yongli, et al. Pattern Recognition of Unknown Partial Discharge Types Based on KELM-VPMCD Method ［J］. Power Automation Equipment, 2018(5):141-147.
［8］TANG T, BO L, LIU X, et al. Variable Predictive Model Class Discrimination Using Novel Predictive Models and Adaptive Feature Selection for Bearing Faultidentification ［J］. Journal of Sound and Vibration, 2018,425:137-148.
［9］郑艳艳, 朱永利, 高佳程. 基于支持向量回归的VPMCD方法及其在局部放电模式识别中的应用［J］. 华北电力大学学报(自然科学版), 2019, 46(2):23-28.
ZHENG Yanyan, ZHU Yongli, GAO Jiacheng. VPMCD Method Based on Support Vector Regression and its Application in Partial Discharge Pattern Recognition［J］. Journal of North China Electric Power University(Natural Science Edition), 2019, 46 (2):23-28.
［10］LIU X, BO L. Identification of Resonance States of Rotor-bearing System Using RQA and Optimal Binary Tree SVM［J］.Neurocomputing, 2015,152:36-44.
［11］KANTZ H, SCHREIBER T. Nonlinear Time Series Analysis［M］.Cambridge:Cambridge University Press, 2004.
［12］KENNEL M B, BROWN R, ABARBANEL H D. Determining Embedding Dimension for Phase-space Reconstruction Using a Geometrical Construction［J］.Physical Review A, 1992, 45(6):3403.
［13］PENG H， LONG F， DING C. Feature Selection Based on Mutual Information:Criteria of Max-dependency, Max-relevance, and Min-redundancy［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2005,27(8):1226-1238.
［14］BISHOP C M. Neural Networks for Pattern Re-cognition［J］. Advances in Computers, 1993, 37:119-166.
［15］HE X, CAI D, NIYOGI P. Laplacian Score for Feature Selection［J］.Advances in Neural Information Processing Systems, 2006,18:507.
［16］ZHANG D, CHEN S, ZHOU Z. Constraint Score:a New Filter Method for Feature Selection with Pairwise Constraints［J］. Pattern Recognition, 2007,41(5)：1440-1451.
［17］ROBNIK-IKONJA M, KONONENKO I. Theoretical and Empirical Analysis of RELIEFF and RRELIEFF［J］. Machine Learning, 2003,53(1):23-69.
［18］STEUER R, KURTHS J, DAUB C O, et al. The Mutual Information:Detecting and Evaluating Dependencies between Variables［J］. Bioinformatics, 2002,18(S):231-240.
［19］VEER L J V T, HONGYUE D, VIJVER M J V D, et al. Gene Expression Profiling Predicts Clinical Outcome of Breast Cancer［J］. Nature, 2002, 415(6871):530.
［20］SHEN W, GUO X, WU C, et al. Forecasting Stock Indices Using Radial Basis Function Neural Networks Optimized by Artificial Fish Swarm Algorithm［J］.Knowledge-based Systems, 2011, 24(3):378-385.
［21］LI H, GUO S, LI C, et al. A Hybrid Annual Power Load-forecasting Model Based on Generalized Regression Neural Network with Fruit Fly Optimization Algorithm［J］.Knowledge-based Systems, 2013, 37:378-387.

[1]	罗洁思, 于德介, 史美丽. 基于SVD和线调频小波路径追踪的转速波动齿轮箱故障诊断 [J]. J4, 201016, 21(16): 1947-1951.
[2]	肖刚, 顾海瑞, 董锦锦, 王琪冰, 陆佳炜. 仿真数据驱动的长期服役电梯导轨故障迁移诊断方法[J]. 中国机械工程, 2024, 35(01): 125-135.
[3]	于树博, 刘占生, 赵辰. 动力学仿真数据驱动的域自适应智能诊断方法[J]. 中国机械工程, 2023, 34(23): 2832-2841.
[4]	陈剑, 许畅, 徐庭亮, . 基于位错叠加法和改进概率神经网络的离心泵故障诊断方法[J]. 中国机械工程, 2023, 34(23): 2854-2861.
[5]	秦国浩, 张楷, 丁昆, 黄锋飞, 郑庆, 丁国富, . 动态宽卷积残差网络的轴承故障诊断方法[J]. 中国机械工程, 2023, 34(18): 2212-2221.
[6]	董绍江, 周存芳, 陈里里, 徐向阳. 基于判别性特征提取和双重域对齐的轴承跨域故障诊断[J]. 中国机械工程, 2023, 34(15): 1856-1863.
[7]	蒙康, 滕伟, 彭迪康, 向玲, 柳亦兵. 运行机理与数据双驱动的风电齿轮箱系统故障预警[J]. 中国机械工程, 2023, 34(12): 1476-1485.
[8]	郑近德, 陈焱, 童靳于, 潘海洋. 精细广义复合多元多尺度反向散布熵及其在滚动轴承故障诊断中的应用[J]. 中国机械工程, 2023, 34(11): 1315-1325.
[9]	张龙, 胡燕青, 赵丽娟, 张号. 多通道信息融合与深度迁移学习的旋转机械故障诊断[J]. 中国机械工程, 2023, 34(08): 966-975.
[10]	张龙, 张号, 王朝兵, 王晓博, 文培田, 赵丽娟. 显式动力学驱动的轴承变工况故障诊断方法[J]. 中国机械工程, 2023, 34(08): 982-992.
[11]	董绍江, 朱朋, 朱孙科, 刘兰徽, 邢镔, 胡小林. 基于仿真数据驱动和领域自适应的滚动轴承故障诊断方法[J]. 中国机械工程, 2023, 34(06): 694-702.
[12]	赵靖, 杨绍普, 李强, 刘永强, . 一种残差注意力迁移学习方法及其在滚动轴承故障诊断中的应用[J]. 中国机械工程, 2023, 34(03): 332-343.
[13]	赵艺珂, 王家序, 张新, 吴磊, 刘治汶. 基于增强最小熵解卷积的航空发动机故障诊断[J]. 中国机械工程, 2023, 34(02): 193-200.
[14]	李可, 燕晗, 顾杰斐, 宿磊, 苏文胜, 薛志钢. 基于shapelets时间序列的多源迁移学习滚动轴承故障诊断方法[J]. 中国机械工程, 2022, 33(24): 2990-2996,3006.
[15]	郭伟, 邢晓松. 基于改进卷积生成对抗网络的少样本轴承智能诊断方法[J]. 中国机械工程, 2022, 33(19): 2347-2355.