基于PCA-ELM的弹载组合导航智能故障检测算法
|
摘要
针对传统PCA-ELM(主元分析-极限学习机)算法分类效果稳定性差和准确率不高等问题,结合弹载组合导航系统卡尔曼滤波器,提出一种改进PCA-ELM故障检测方法。首先,分析了PCA算法负载矩阵与卡尔曼滤波新息协方差矩阵的关系,构造新的权系数矩阵,并引入极限学习机对权系数矩阵进行参数优化,将参数优化后的负载矩阵进行故障分析。最后,将该算法首次应用于弹载组合导航系统。仿真实验表明,在检测斜坡型故障方面,检测速度和检测正确率均优于传统PCA,MSS(多子集分离法)及AIME(自主完好性外推法)算法。
In view of the problem of poor stability and low accuracy in the traditional PCA-ELM (principal component analysis-extreme learning machine) algorithm,an improved PCA-ELM fault detection method is proposed by combining the Kalman filter of the missile borne integrated navigation system.Firstly,the relationship between the load matrix of the PCA algorithm and the covariance matrix of the Kalman filter is analyzed. Then,the new weight coefficient matrix is introduced,and the extreme learning machine is introduced to optimize the weight coefficient matrix,and the load matrix is optimized for fault analysis. Finally,the algorithm is applied to the missile borne integrated navigation system for the first time. The simulation experiment shows that the method proposed in this paper is superior to the traditional PCA,MSS and AIME algorithms in the detection speed and detection accuracy.
In view of the problem of poor stability and low accuracy in the traditional PCA-ELM (principal component analysis-extreme learning machine) algorithm,an improved PCA-ELM fault detection method is proposed by combining the Kalman filter of the missile borne integrated navigation system.Firstly,the relationship between the load matrix of the PCA algorithm and the covariance matrix of the Kalman filter is analyzed. Then,the new weight coefficient matrix is introduced,and the extreme learning machine is introduced to optimize the weight coefficient matrix,and the load matrix is optimized for fault analysis. Finally,the algorithm is applied to the missile borne integrated navigation system for the first time. The simulation experiment shows that the method proposed in this paper is superior to the traditional PCA,MSS and AIME algorithms in the detection speed and detection accuracy.
引文
[1]张威,魏炳翌,闻新.国内航天器故障诊断技术应用状况分析与展望[J].航空兵器,2017(4):66-74.Zhang Wei,Wei Bingyi,Wen Xin.Application Status A-nalysis and Prospect on Spacecraft Fault Diagnosis Technology in China[J].Aero Weaponry,2017(4):66-74.(in Chinese)
[2]高为广,杨元喜.神经网络辅助的GPS/INS组合导航故障检测算法[J].测绘学报,2008,37(4):403-409.Gao Weiguang,Yang Yuanxi.Neural Network Aided GPS/INS Integrated Navigation Fault Detection Algorithms[J].Acta Geodaetica et Cartographica Sinica,2008,37(4):403-409.(in Chinese)
[3]连远锋,李国和,吴发林,等.基于遗传PNN网络的组合导航故障诊断研究[J].仪器仪表学报,2012,33(1):120-126.Lian Yuanfeng,Li Guohe,Wu Falin,et al.Fault-Diagnosis Method for INS/GPS Integrated Navigation System Based on PNN and Genetic Algorithm[J].Chinese Journal of Scientific Instrument,2012,33(1):120-126.(in Chinese)
[4]程洪炳,倪世宏,黄国荣,等.UKF强跟踪滤波在组合导航故障诊断中的应用[J].压电与声光,2011,33(6):901-905.Cheng Hongbing,Ni Shihong,Huang Guorong,et al.Application of UKF Strong Tracking Filter in Fault Detection of Integrated Navigation System[J].Piezoelectrics&Acoustooptics,2011,33(6):901-905.(in Chinese)
[5]杨长林,欧迪栋,刘延飞.基于联邦滤波的组合导航故障诊断与容错研究[J].科学技术与工程,2012,12(32):8601-8605.Yang Changlin,Ou Didong,Liu Yanfei.Research on Fault Detection and Fault-Tolerant for Integrated Navigation System Based on Federated Kalman Filter Method[J].Science Technology and Engineering,2012,12(32):8601-8605.(in Chinese)
[6]阮雯,秦红磊,丛丽,等.改进残差χ2检验法在组合导航故障检测中应用[J].电子测量技术,2012,35(11):124-128.Ruan Wen,Qin Honglei,Cong Li,et al.Improved Residualχ2Inspection in Integrated Navigation Fault Detection of Application[J].Electronic Measurement Technology,2012,35(11):124-128.(in Chinese)
[7]Bhatti U I,Ochieng W Y.Failure Modes and Models for Integrated GPS/INS Systems[J].Journal of Navigation,2007,60(2):327-348.
[8]蒋德杰,岳亚洲,刘海颖,等.基于速率检测的GNSS/INS快速完好性监测方法[J].航空兵器,2012(4):11-14.Jiang Dejie,Yue Yazhou,Liu Haiying,et al.Fast Integrity Monitoring Method for the GNSS/INS System Based on Rate Detection[J].Aero Weaponry,2012(4):11-14.(in Chinese)
[9]陈绍炜,吴敏华,赵帅.基于PCA和ELM的模拟电路故障诊断[J].计算机工程与应用,2015,51(11):248-251.Chen Shaowei,Wu Minhua,Zhao Shuai.Analog Circuit Fault Diagnosis Based on PCA and ELM[J].Computer Engineering and Applications,2015,51(11):248-251.(in Chinese)
[10]闫帅领,郑静静,刘光展.船舶组合导航系统故障识别的神经网络方法[J].舰船科学技术,2017,39(11):106-108.Yan Shuailing,Zheng Jingjing.Liu Guangzhan.Neural Network Method for Fault Identification of Marine Integrated Navigation System[J].Ship Science and Technology,2017,39(11):106-108.(in Chinese)
[11]陈绍炜,柳光峰,冶帅,等.基于蝙蝠算法优化ELM的模拟电路故障诊断研究[J].电子测量技术,2015,38(2):138-141.Chen Shaowei,Liu Guangfeng,Ye Shuai,et al.Research for Analog Circuit Fault Diagnosis Based on ELMOptimized by Bat Algorithm[J].Electronic Measurement Technology,2015,38(2):138-141.(in Chinese)
[12]裘日辉,刘康玲,谭海龙,等.基于极限学习机的分类算法及在故障识别中的应用[J].浙江大学学报:工学版,2016,50(10):1965-1972.Qiu Rihui,Liu Kangling,Tan Hailong,et al.Classification Algorithm Based on Extreme Learning Machine and its Application in Fault Identification of Tennessee Eastman Process[J].Journal of Zhejiang University:Engineering Science,2016,50(10):1965-1972.(in Chinese)
[13]陈楚瑶,朱大奇.神经网络主元分析的传感器故障诊断方法[J].系统工程与电子技术,2010,32(7):1549-1553.Chen Chuyao,Zhu Daqi.Sensor Fault Diagnosis Method Based on Neural Network Principal Component Analysis[J].Systems Engineering and Electronics,2010,32(7):1549-1553.(in Chinese)
[14]周东华,李钢,李元.数据驱动的工业过程故障诊断技术[M].北京:科学出版社,2011:25-27.Zhou Donghua,Li Gang,Li Yuan.Data Driven Fault Diagnosis Technology for Industrial Process[M].Beijing:Science Press,2011:25-27.(in Chinese)
[15]秦永元.惯性导航[M].北京:科学出版社,2006:355-360.Qing Yongyuan.Inertial Navigation[M].Beijing:Science Press,2006:355-360.(in Chinese)
[2]高为广,杨元喜.神经网络辅助的GPS/INS组合导航故障检测算法[J].测绘学报,2008,37(4):403-409.Gao Weiguang,Yang Yuanxi.Neural Network Aided GPS/INS Integrated Navigation Fault Detection Algorithms[J].Acta Geodaetica et Cartographica Sinica,2008,37(4):403-409.(in Chinese)
[3]连远锋,李国和,吴发林,等.基于遗传PNN网络的组合导航故障诊断研究[J].仪器仪表学报,2012,33(1):120-126.Lian Yuanfeng,Li Guohe,Wu Falin,et al.Fault-Diagnosis Method for INS/GPS Integrated Navigation System Based on PNN and Genetic Algorithm[J].Chinese Journal of Scientific Instrument,2012,33(1):120-126.(in Chinese)
[4]程洪炳,倪世宏,黄国荣,等.UKF强跟踪滤波在组合导航故障诊断中的应用[J].压电与声光,2011,33(6):901-905.Cheng Hongbing,Ni Shihong,Huang Guorong,et al.Application of UKF Strong Tracking Filter in Fault Detection of Integrated Navigation System[J].Piezoelectrics&Acoustooptics,2011,33(6):901-905.(in Chinese)
[5]杨长林,欧迪栋,刘延飞.基于联邦滤波的组合导航故障诊断与容错研究[J].科学技术与工程,2012,12(32):8601-8605.Yang Changlin,Ou Didong,Liu Yanfei.Research on Fault Detection and Fault-Tolerant for Integrated Navigation System Based on Federated Kalman Filter Method[J].Science Technology and Engineering,2012,12(32):8601-8605.(in Chinese)
[6]阮雯,秦红磊,丛丽,等.改进残差χ2检验法在组合导航故障检测中应用[J].电子测量技术,2012,35(11):124-128.Ruan Wen,Qin Honglei,Cong Li,et al.Improved Residualχ2Inspection in Integrated Navigation Fault Detection of Application[J].Electronic Measurement Technology,2012,35(11):124-128.(in Chinese)
[7]Bhatti U I,Ochieng W Y.Failure Modes and Models for Integrated GPS/INS Systems[J].Journal of Navigation,2007,60(2):327-348.
[8]蒋德杰,岳亚洲,刘海颖,等.基于速率检测的GNSS/INS快速完好性监测方法[J].航空兵器,2012(4):11-14.Jiang Dejie,Yue Yazhou,Liu Haiying,et al.Fast Integrity Monitoring Method for the GNSS/INS System Based on Rate Detection[J].Aero Weaponry,2012(4):11-14.(in Chinese)
[9]陈绍炜,吴敏华,赵帅.基于PCA和ELM的模拟电路故障诊断[J].计算机工程与应用,2015,51(11):248-251.Chen Shaowei,Wu Minhua,Zhao Shuai.Analog Circuit Fault Diagnosis Based on PCA and ELM[J].Computer Engineering and Applications,2015,51(11):248-251.(in Chinese)
[10]闫帅领,郑静静,刘光展.船舶组合导航系统故障识别的神经网络方法[J].舰船科学技术,2017,39(11):106-108.Yan Shuailing,Zheng Jingjing.Liu Guangzhan.Neural Network Method for Fault Identification of Marine Integrated Navigation System[J].Ship Science and Technology,2017,39(11):106-108.(in Chinese)
[11]陈绍炜,柳光峰,冶帅,等.基于蝙蝠算法优化ELM的模拟电路故障诊断研究[J].电子测量技术,2015,38(2):138-141.Chen Shaowei,Liu Guangfeng,Ye Shuai,et al.Research for Analog Circuit Fault Diagnosis Based on ELMOptimized by Bat Algorithm[J].Electronic Measurement Technology,2015,38(2):138-141.(in Chinese)
[12]裘日辉,刘康玲,谭海龙,等.基于极限学习机的分类算法及在故障识别中的应用[J].浙江大学学报:工学版,2016,50(10):1965-1972.Qiu Rihui,Liu Kangling,Tan Hailong,et al.Classification Algorithm Based on Extreme Learning Machine and its Application in Fault Identification of Tennessee Eastman Process[J].Journal of Zhejiang University:Engineering Science,2016,50(10):1965-1972.(in Chinese)
[13]陈楚瑶,朱大奇.神经网络主元分析的传感器故障诊断方法[J].系统工程与电子技术,2010,32(7):1549-1553.Chen Chuyao,Zhu Daqi.Sensor Fault Diagnosis Method Based on Neural Network Principal Component Analysis[J].Systems Engineering and Electronics,2010,32(7):1549-1553.(in Chinese)
[14]周东华,李钢,李元.数据驱动的工业过程故障诊断技术[M].北京:科学出版社,2011:25-27.Zhou Donghua,Li Gang,Li Yuan.Data Driven Fault Diagnosis Technology for Industrial Process[M].Beijing:Science Press,2011:25-27.(in Chinese)
[15]秦永元.惯性导航[M].北京:科学出版社,2006:355-360.Qing Yongyuan.Inertial Navigation[M].Beijing:Science Press,2006:355-360.(in Chinese)