传感器故障后多变量经验小波变换多点预测
|
摘要
为有效应对多点风速传感器或风压传感器故障而造成的损失,同时为了降低运算的复杂性和工程应用的难度,需要提出同步恢复缺失数据的模型。传统的多通道信号诊断采用多元经验模态分解(multivariate empiricalmode decomposition,简称MEMD),笔者提出多变量经验小波变换(multivariable empirical wavelet transform,简称MEWT)来同步恢复多点缺失数据。具体应用时,首先,运用MEWT将多点信号同时分解为一系列模态;然后,利用核函数极限学习机(kernel-based extreme learning machine,简称KELM)实现同步预测,同时运用杜鹃搜索(cuckoo search,简称CS)算法对模型的正则化参数以及核参数进行智能寻优。多步预测时,采用多输入多输出(multi-input multi-output,简称MIMO)策略代替传统的滚动策略。建筑物表面实测多点风压数据和实测多点下击暴流风速数据用于验证模型的可行性。与噪声辅助的多元经验模态分解核函数极限学习机的对比结果表明,该模型能更高精度地同步恢复多点多步信号。
In order to effectively decrease the loss caused by the multipoint fault of wind speed sensors or wind pressure sensors,and to reduce the complexity of computation and the difficulty of the engineering application,a model needs to be proposed to recover the missing data at the same time.As the traditional multi-channel signal diagnosis uses multivariate empirical mode decomposition(MEMD),the multivariable empirical wavelet transform(MEWT)is proposed to restore the multipoint missing data synchronously.In practical application,the multipoint signals are decomposed into a series of modes at the same time,and then the kernel-based extreme learning machine(KELM)is used to predict,and the cuckoo search(CS)algorithm is used to optimize the regularization parameters of the model and the kernel parameters.For multi-step forecasting,the traditional recursive strategy is replaced by the multiple-input multiple-output(MIMO)strategy.The actual measured multipoint wind pressure on the building surface and the measured multipoint data of the downburst are used to verify the feasibility of the model.Compared with the noise assisted multivariate empirical mode decomposition kernel-based extreme learning machine(NA-MEMD-KELM-CS),the result shows that the proposed model can recover signals simultaneously with high accuracy.
In order to effectively decrease the loss caused by the multipoint fault of wind speed sensors or wind pressure sensors,and to reduce the complexity of computation and the difficulty of the engineering application,a model needs to be proposed to recover the missing data at the same time.As the traditional multi-channel signal diagnosis uses multivariate empirical mode decomposition(MEMD),the multivariable empirical wavelet transform(MEWT)is proposed to restore the multipoint missing data synchronously.In practical application,the multipoint signals are decomposed into a series of modes at the same time,and then the kernel-based extreme learning machine(KELM)is used to predict,and the cuckoo search(CS)algorithm is used to optimize the regularization parameters of the model and the kernel parameters.For multi-step forecasting,the traditional recursive strategy is replaced by the multiple-input multiple-output(MIMO)strategy.The actual measured multipoint wind pressure on the building surface and the measured multipoint data of the downburst are used to verify the feasibility of the model.Compared with the noise assisted multivariate empirical mode decomposition kernel-based extreme learning machine(NA-MEMD-KELM-CS),the result shows that the proposed model can recover signals simultaneously with high accuracy.
引文
[1]Rehman N,Mandic D P.Multivariate empirical mode decomposition[J].Proceedings Mathematical Physical&Engineering Sciences,2010,466(2117):1291-1302.
[2]LüYong,Yuan Rui,Song Gangbing.Multivariate empirical mode decomposition and its application to fault diagnosis of rolling bearing[J].Mechanical Systems&Signal Processing,2016,81:219-234.
[3]Huang Weiping,Zeng Jing,Wang Ziyang,et al.Partial noise assisted multivariate EMD:an improved noise assisted method for multivariate signals decomposition[J].Biomedical Signal Processing&Control,2017,36:205-220.
[4]熊炘,杨世锡,甘春标,等.齿轮点蚀的多通道数据融合识别方法[J].振动、测试与诊断,2014,34(1):63-68.Xiong Xin,Yang Shixi,Gan Chunbiao,et al.Multichannel data fusion for the identification of gear pitting[J].Journal of Vibration,Measurement&Diagnosis,2014,34(1):63-68.(in Chinese)
[5]王恒,杜世昌,奚立峰,等.基于改进多元经验模态分解的多通道振动信号融合分析[J].机械设计与研究,2018(2):31-36.Wang Heng,Du Shichang,Xi Lifeng,et al.Multichannel vibration signal analysis based onimproved multivariate empirical mode decomposition[J].Machine Design&Research,2018(2):31-36.(in Chinese)
[6]段若晨,王丰华,周荔丹,等.利用窄带噪声辅助多元经验模态分解算法检测换流变压器用有载分接开关机械状态[J].电工技术学报,2017,32(10):182-189.Duan Ruochen,Wang Fenghua,Zhou Lidan,et al.Mechanical condition detection of on-load tap-changer in converter transformer based on narrowband noise assisted multivariate empirical mode decomposition algorithm[J].Transactions of China Electrotechnical Society,2017,32(10):182-189.(in Chinese)
[7]Zhang Chu,Zhou Jianzhou,Li Chaoshun,et al.Acompound structure of ELM based on feature selection and parameter optimization using hybrid backtracking search algorithm for wind speed forecasting[J].Energy Conversion&Management,2017,143:360-376.
[8]Gilles J.Empirical wavelet transform[J].IEEETransactions on Signal Processing,2013,61(16):3999-4010.
[9]Gilles J,Heal K.A parameterless scale-space approach to find meaningful modes in histograms u2014Application to image and spectrum segmentation[J].International Journal of Wavelets Multiresolution&Information Processing,2014,12(6):191-209.
[10]Huang Gao,Huang Guangbin,Song Shiji,et al.Trends in extreme learning machines:a review[J].Neural Networks the Official Journal of the International Neural Network Society,2015,61(C):32.
[11]Cheng Chi,Tay W P,Huang Guangbin.Extreme learning machines for intrusion detection[C]∥International Joint Conference on Neural Networks.[S.l.]:IEEE,2012:1-8.
[12]Yang Xinshe,Deb S.Cuckoo search:recent advances and applications[J].Neural Computing&Applications,2014,24(1):169-174.
[13]Zhang Kequan,Qu Zongxi,Wang Jianzhou,et al.Anovel hybrid approach based on cuckoo search optimization algorithm for short‐term wind speed forecasting[J].Environmental Progress&Sustainable Energy,2017,36(3):943-952.
[14]叶志伟,赵伟,王明威,等.一种基于杜鹃搜索算法的图像自适应增强方法[J].测绘科学技术学报,2016,33(1):38-42.Ye Zhiwei,Zhao Wei,Wang Mingwei,et al.An adaptive image enhancement approach based on cuckoo search algorithm[J].Journal of Geomatics Science and Technology,2016,33(1):38-42.(in Chinese)
[15]Civicioglu P,Besdok E.A conceptual comparison of the Cuckoo-search,particle swarm optimization,differential evolution and artificial bee colony algorithms[J].Artificial Intelligence Review,2013,39(4):315-346.
[16]Wong P K,Wong K I,Vong C M,et al.Modeling and optimization of biodiesel engine performance using kernel-based extreme learning machine and cuckoo search[J].Renewable Energy,2015,74:640-647.
[17]Taieb S B,Bontempi G,Atiya A F,et al.A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5forecasting competition[J].Expert Systems with Applications,2012,39(8):7067-7083.
[18]Zhao Jing,Wang Jianzhou,Liu Feng.Multistep forecasting for short-term wind speed using an optimized extreme learning machine network with decompositionbased signal filtering[J].Journal of Energy Engineering,2016,142(3):04015036-1-21.
[19]Yang Dixiong,Yang Kaisheng.Multi-step prediction of strong earthquake ground motions and seismic responses of SDOF systems based on EMD-ELM method[J].Soil Dynamics&Earthquake Engineering,2016,85:117-129.
[20]李锦华,吴春鹏,陈水生.矩形结构非高斯风荷载特性研究[J].振动、测试与诊断,2014,34(5):951-959.Li Jinhua,Wu Chunpeng,Chen Shuisheng.Characteristics of non-Gaussian wind pressures on rectangular structure[J].Journal of Vibration,Measurement&Diagnosis,2014,34(5):951-959.(in Chinese)
[2]LüYong,Yuan Rui,Song Gangbing.Multivariate empirical mode decomposition and its application to fault diagnosis of rolling bearing[J].Mechanical Systems&Signal Processing,2016,81:219-234.
[3]Huang Weiping,Zeng Jing,Wang Ziyang,et al.Partial noise assisted multivariate EMD:an improved noise assisted method for multivariate signals decomposition[J].Biomedical Signal Processing&Control,2017,36:205-220.
[4]熊炘,杨世锡,甘春标,等.齿轮点蚀的多通道数据融合识别方法[J].振动、测试与诊断,2014,34(1):63-68.Xiong Xin,Yang Shixi,Gan Chunbiao,et al.Multichannel data fusion for the identification of gear pitting[J].Journal of Vibration,Measurement&Diagnosis,2014,34(1):63-68.(in Chinese)
[5]王恒,杜世昌,奚立峰,等.基于改进多元经验模态分解的多通道振动信号融合分析[J].机械设计与研究,2018(2):31-36.Wang Heng,Du Shichang,Xi Lifeng,et al.Multichannel vibration signal analysis based onimproved multivariate empirical mode decomposition[J].Machine Design&Research,2018(2):31-36.(in Chinese)
[6]段若晨,王丰华,周荔丹,等.利用窄带噪声辅助多元经验模态分解算法检测换流变压器用有载分接开关机械状态[J].电工技术学报,2017,32(10):182-189.Duan Ruochen,Wang Fenghua,Zhou Lidan,et al.Mechanical condition detection of on-load tap-changer in converter transformer based on narrowband noise assisted multivariate empirical mode decomposition algorithm[J].Transactions of China Electrotechnical Society,2017,32(10):182-189.(in Chinese)
[7]Zhang Chu,Zhou Jianzhou,Li Chaoshun,et al.Acompound structure of ELM based on feature selection and parameter optimization using hybrid backtracking search algorithm for wind speed forecasting[J].Energy Conversion&Management,2017,143:360-376.
[8]Gilles J.Empirical wavelet transform[J].IEEETransactions on Signal Processing,2013,61(16):3999-4010.
[9]Gilles J,Heal K.A parameterless scale-space approach to find meaningful modes in histograms u2014Application to image and spectrum segmentation[J].International Journal of Wavelets Multiresolution&Information Processing,2014,12(6):191-209.
[10]Huang Gao,Huang Guangbin,Song Shiji,et al.Trends in extreme learning machines:a review[J].Neural Networks the Official Journal of the International Neural Network Society,2015,61(C):32.
[11]Cheng Chi,Tay W P,Huang Guangbin.Extreme learning machines for intrusion detection[C]∥International Joint Conference on Neural Networks.[S.l.]:IEEE,2012:1-8.
[12]Yang Xinshe,Deb S.Cuckoo search:recent advances and applications[J].Neural Computing&Applications,2014,24(1):169-174.
[13]Zhang Kequan,Qu Zongxi,Wang Jianzhou,et al.Anovel hybrid approach based on cuckoo search optimization algorithm for short‐term wind speed forecasting[J].Environmental Progress&Sustainable Energy,2017,36(3):943-952.
[14]叶志伟,赵伟,王明威,等.一种基于杜鹃搜索算法的图像自适应增强方法[J].测绘科学技术学报,2016,33(1):38-42.Ye Zhiwei,Zhao Wei,Wang Mingwei,et al.An adaptive image enhancement approach based on cuckoo search algorithm[J].Journal of Geomatics Science and Technology,2016,33(1):38-42.(in Chinese)
[15]Civicioglu P,Besdok E.A conceptual comparison of the Cuckoo-search,particle swarm optimization,differential evolution and artificial bee colony algorithms[J].Artificial Intelligence Review,2013,39(4):315-346.
[16]Wong P K,Wong K I,Vong C M,et al.Modeling and optimization of biodiesel engine performance using kernel-based extreme learning machine and cuckoo search[J].Renewable Energy,2015,74:640-647.
[17]Taieb S B,Bontempi G,Atiya A F,et al.A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5forecasting competition[J].Expert Systems with Applications,2012,39(8):7067-7083.
[18]Zhao Jing,Wang Jianzhou,Liu Feng.Multistep forecasting for short-term wind speed using an optimized extreme learning machine network with decompositionbased signal filtering[J].Journal of Energy Engineering,2016,142(3):04015036-1-21.
[19]Yang Dixiong,Yang Kaisheng.Multi-step prediction of strong earthquake ground motions and seismic responses of SDOF systems based on EMD-ELM method[J].Soil Dynamics&Earthquake Engineering,2016,85:117-129.
[20]李锦华,吴春鹏,陈水生.矩形结构非高斯风荷载特性研究[J].振动、测试与诊断,2014,34(5):951-959.Li Jinhua,Wu Chunpeng,Chen Shuisheng.Characteristics of non-Gaussian wind pressures on rectangular structure[J].Journal of Vibration,Measurement&Diagnosis,2014,34(5):951-959.(in Chinese)