基于HMM校正与神经网络延拓的EMD端点效应抑制方法

中国机械工程 ›› 2015, Vol. 26 ›› Issue (14): 1920-1925.

基于HMM校正与神经网络延拓的EMD端点效应抑制方法

孟宗1,2;闫晓丽1;王赛3

1.河北省测试计量技术及仪器重点实验室(燕山大学)，秦皇岛，066004
2.国家冷轧板带装备及工艺工程技术研究中心，秦皇岛，066004
3.长城汽车股份有限公司技术中心(河北省汽车工程技术研究中心)，保定，071000

出版日期:2015-07-25 发布日期:2015-08-05
基金资助:
国家自然科学基金资助项目(51105323)；河北省自然科学基金资助项目(E2015203356)；河北省高等学校科学研究计划资助重点项目(ZD2015049)

Restraining Method of End Effect for EDM Based on Error Calibration by HMM and Neural Network

Meng Zong1,2;Yan Xiaoli1;Wang Sai3

1.Key Laboratory of Measurement Technology and Instrumentation of Hebei Province,Yanshan University,Qinhuangdao,Hebei,066004
2.National Engineering Research Center for Equipment and Technology of Cold Rolling Strip,Qinhuangdao,Hebei,066004
3.R&D Center of Great Wall Motor Company(Automotive Engineering Technical Center of Hebei),Baoding,Hebei,071000

Online:2015-07-25 Published:2015-08-05
Supported by:
National Natural Science Foundation of China（No. 51105323）;Hebei Provincial Natural Science Foundation of China（No. E2015203356）

摘要/Abstract

摘要：

针对神经网络延拓方法在抑制经验模态分解的端点效应时存在的延拓数据与真实数据往往存在误差的问题，提出了一种基于HMM校正的方法来减小预测延拓数据误差。首先利用径向基函数(RBF)神经网络预测估计方法对部分原始数据进行估计，同时对端点外数据进行预测。然后计算该方法估计的数据与真实数据的误差序列,再用HMM方法建立估计误差序列模型，用以预测延拓后数据的误差。最后用RBF神经网络延拓数据减去HMM预测的误差数据得到新的校正后延拓数据。仿真与实验证明了将HMM预测方法与RBF神经网络数据延拓结合应用到解决端点效应的过程中所得到的延拓数据更接近真实数据，能够更好地解决端点效应问题，提高了经验模态分解精度。

关键词: 隐马尔科夫模型, 误差校正, 神经网络, 端点效应, 经验模态分解

Abstract:

End effects reduced the precision of EMD greatly, and neural network extension was used to restrain the end effects. However, there were forecast errors in the data forecasted by neural network. Here, a new method, error calibration by HMM was proposed to solve the problem. The radial basis function(RBF) neural network was firstly used to forecast outboard of both ends and estimation part of original signals. Then HMM was used to analyze the forecasting errors and to find the regularity of forecast errors of neural network. According to the analysis, the forecasting errors in next step were analyzed and forecasted to adjust neural network's forecasting results. Simulation and experiments prove that the method can solve the end effects effectively.

Key words: hidden Markov model(HMM), error calibration, neural network, end effect, empirical mode decomposition (EMD)

中图分类号:

孟宗, 闫晓丽, 王赛. 基于HMM校正与神经网络延拓的EMD端点效应抑制方法[J]. 中国机械工程, 2015, 26(14): 1920-1925.

Meng Zong, Yan Xiaoli, Wang Sai. Restraining Method of End Effect for EDM Based on Error Calibration by HMM and Neural Network[J]. China Mechanical Engineering, 2015, 26(14): 1920-1925.

参考文献

[1]Huang N E,Shen Z,Long S R,et al.The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis[J].Proceedings of Royal Society of London A, 1998,454:903-995.
[2]Huang W,Shen Z,Huang N E,et al.Nonlinear Indicial Response of Complex Nonstationary Oscillations as Pulmonary Hypertension Responding to Step Hypoxia[J].Proceedings of the National Academy of Science, USA, 1999,96(5):1834-1839.
[3]杨宇,于德介,程军圣.基于EMD与神经网络的滚动轴承故障诊断方法[J].振动与冲击,2005,24(1): 85-88.
Yang Yu,Yu Dejie,Cheng Junsheng.Roller Bearing Fault Diagnosis Method Based on EMD and Neural Network[J].Journal of Vibration and Shock,2005,24(1): 85-88.
[4]程军圣,于德介,杨宇.基于EMD和分形维数的转子系统故障诊断[J].中国机械工程,2005,16(12):1088-1091.
Cheng Junsheng,Yu Dejie,Yang Yu.Fault Diagnosis for Rotor System Based on EMD and Fractal Dimension[J].China Mechanical Engineering,2005,16(12):1088-1091.
[5]孟宗,李姗姗.基于小波半软阈值和EMD的旋转机械故障诊断[J].中国机械工程, 2013, 24(10): 1279-1283.
Meng Zong,Li Shanshan.Research on Fault Diagnosis for Rotating Machinery Based on Semi-soft Wavelet Threshold and EMD[J]. China Mechanical Engineering,2013,24(10): 1279-1283.
[6]陈忠,郑时雄.EMD信号分析方法边缘效应分析[J].数据采集与处理, 2003, 18(1): 114-118.
Chen Zhong,Zheng Shixiong.Analysis on End Effects of EMD Method[J].Journal of Data Acquisition & Processing, 2003, 18(1): 114-118.
[7]杜陈艳,张瑜锋,杨平.经验模态分解边缘效应抑制方法综述[J].仪器仪表学报,2009,30(1):55-60.
Du Chenyan,Zhang Yufeng,Yang Ping,et al.Approaches for the End Effect Restraint of Empirical Mode Decomposition Algorithm[J]. Chinese Journal of Scientific Instrument,2009,30(1):55-60.
[8]胡劲松,杨世锡.EMD方法基于AR模型预测的数据延拓与应用[J].振动、测试与诊断,2007,27(2):116-120.
Hu Jingsong, Yang Shixi.AR Model Prediction-based EMD Method and Its Application to Data Extension[J].Journal of Vibration, Measurement & Diagnosis,2007,27(2):116-120.
[9]韩建平,钱炯,董小军.采用镜像延拓和RBF神经网络处理EMD中端点效应[J].振动、测试与诊断,2010,30(4)：414-417.
Han Jianping,Qian Jiong,Dong Xiaojun.Suppression of End-effect in Empirical Mode Decomposition by Mirror Extension and Radial Basis Function Neural Network Prediction[J].Journal of Vibration,Measurement & Diagnosis,2010,30(4):414-417.
[10]程军圣，于德介，杨宇.基于支持矢量回归机的Hilbert-Huang变换端点效应问题的处理方法[J].机械工程学报,2006,42(4):23-30.
Cheng Junsheng, Yu Dejie, Yang Yu.Process Method for End Effects of Hilbert-Huang Transform Based on Support Vector Regression Machines[J]. Chinese Journal of Mechanical Engineering,2006,42(4):23-30.
[11]孟宗,顾海燕,李姗姗.基于神经网络集成的B样条经验模态分解端点效应抑制方法[J].机械工程学报,2013, 49(9): 106-112.
Meng Zong, Gu Haiyan,Li Shanshan.Restraining Method for End Effect of B-spline Empirical Mode Decomposition Based on Neural Network Ensemble[J].Chinese Journal of Mechanical Engineering,2013,49(9): 106-112.
[12]柳新民,刘冠军,邱静.基于HMM-SVM的故障诊断模型及应用[J].仪器仪表学报,2006,27(1):45-48.
Liu Xinmin,Liu Guanjun,Qiu Jing.Hybrid HMM and SVM Approach for Fault Diagnosis[J].Chinese Journal of Scientific Instrument, 2006, 27(1): 45-48.

[1]	李小力, 陈威, 闫蓉. 基于BP神经网络的空间轮廓误差自适应补偿 [J]. J4, 201016, 21(16): 1902-1906.
[2]	赵伟, 王志远, 周志立. 载重汽车连续下坡弯道路段制动稳定性仿真研究 [J]. J4, 201016, 21(16): 2010-2015.
[3]	陈换过, 刘培君, 俞杭, 肖雪. 基于改进PCA-SOM的电静压伺服作动器油滤堵塞故障诊断[J]. 中国机械工程, 2021, 32(07): 799-805.
[4]	王玉龙, 裴锋, 刘文如, 闫春香, 周卫林, 李智. 基于开关式深度神经网络的拟人化自动驾驶决策算法[J]. 中国机械工程, 2021, 32(06): 689-696.
[5]	李冰1;张永德1;袁立鹏2;朱光强3;代雪松1;苏文海3. 液压四足机器人足端的力预测控制与运动平稳性[J]. 中国机械工程, 2021, 32(05): 523-532.
[6]	郭保苏；庄集超；章钦；吴凤和. 基于多分量卷积神经网络的多晶硅晶片颜色差异检测[J]. 中国机械工程, 2021, 32(03): 297-304.
[7]	钟百鸿；王琳；钟诗胜. 基于综合灰关联序模型的残差门控循环神经网络位标器零部件选配[J]. 中国机械工程, 2021, 32(03): 314-320,356.
[8]	刘晏宇;喻凡;宋娟娟;庞纪苏;KAKU Chuyo. 采用神经网络与模糊控制的制动需求识别[J]. 中国机械工程, 2020, 31(23): 2847-2855.
[9]	于扬;梁军;陈龙;陈小波;朱宁;华国栋. 基于高斯混合隐马尔科夫模型与人工神经网络的紧急换道行为预测方法[J]. 中国机械工程, 2020, 31(23): 2874-2882,2890.
[10]	李锋1;程阳洋1,2;陈勇1;汤宝平3. 用于旋转机械状态趋势预测的量子注意力循环编码解码神经网络[J]. 中国机械工程, 2020, 31(21): 2573-2582.
[11]	曾大懿1;杨基宏2;邹益胜1;张继冬1;宋小欣1. 基于并行多通道卷积长短时记忆网络的轴承寿命预测方法[J]. 中国机械工程, 2020, 31(20): 2454-2462,2471.
[12]	童哲铭1；陈尧1；童水光1；余跃1；李进富2；郝国帅3. 基于NSGA-Ⅲ算法的低比转速离心泵多目标优化设计[J]. 中国机械工程, 2020, 31(18): 2239-2246.
[13]	何彦1;凌俊杰1;王禹林2;李育锋1;吴鹏程1;肖圳1. 基于长短时记忆卷积神经网络的刀具磨损在线监测模型[J]. 中国机械工程, 2020, 31(16): 1959-1967.
[14]	吕佑龙;许鸿伟;郑城;张洁;郑鹏. 基于混合式特征选择模型的晶圆允收测试关键参数识别方法[J]. 中国机械工程, 2020, 31(16): 1978-1984,1997.
[15]	余石龙, 鲍劲松, 李婕, 张启华. 基于误差注意力的晶圆制造数据异常检测[J]. 中国机械工程, 2020, 31(14): 1686-1692.