飞行器惯性导航陀螺仪故障诊断研究
|
摘要
飞行器惯性导航系统中陀螺仪易受工作环境影响,出现性能精度下降等故障问题。为了达到更精确和可靠的陀螺仪故障诊断,提出改进的经验模态分解-排列熵算法(ensemble empirical mode decomposition-permutation entropy,EEMD-PE)的故障诊断方法。应用EEMD自适应分解陀螺仪输出信号为固有模态函数(IMF),利用排列熵对信号微变具有高敏感性的特点,对分解的IMF信号重新构造,重构后的特征向量作为训练集,利用具有快速学习能力和高准确率的概率神经网络(PNN)模型,建立陀螺仪故障诊断网络,训练好的概率神经网络就可以对陀螺仪工作状态进行诊断。将基于EEMD的排列熵故障诊断、基于EEMD的信息熵故障诊断以及基于EMD的排列熵故障诊断效果对比。研究表明:基于EEMD和排列熵算法的诊断方法对故障状态具有更高的辨识能力,更高的精确度,能够作为在线检测陀螺仪故障的有效工具。
In the inertial navigation system of the aircraft, the gyroscope is easy to be affected by working environment, some failure problems such as performance degradation are appeared. An improved empirical mode decomposition permutation entropy algorithm(ensemble empirical mode decomposition-permutation entropy EEMD-PE) is presented for fault diagnosis. EEMD was used to adaptively decompose gyroscope signal into intrinsic mode function(IMF), the IMF signal was reconstructed using the permutation entropy, which has the characteristics of highly sensitivity to the signal micro variation, and the reconstructed eigenvectors are used as training sets. A probabilistic neural network(PNN) model with fast learning ability and high accuracy was used to build a gyroscope fault diagnosis network, and the trained probabilistic neural network can diagnose the working state of the gyroscope. The EEMD based permutation entropy fault diagnosis, the EEMD based information entropy fault diagnosis, and the EMD based permutation entropy fault diagnosis effect were compared. The research shows that the diagnosis method based on EEMD and permutation entropy algorithm has higher identification ability and higher accuracy for fault state, and can be used as an effective tool for on-line detecting gyroscope faults.
In the inertial navigation system of the aircraft, the gyroscope is easy to be affected by working environment, some failure problems such as performance degradation are appeared. An improved empirical mode decomposition permutation entropy algorithm(ensemble empirical mode decomposition-permutation entropy EEMD-PE) is presented for fault diagnosis. EEMD was used to adaptively decompose gyroscope signal into intrinsic mode function(IMF), the IMF signal was reconstructed using the permutation entropy, which has the characteristics of highly sensitivity to the signal micro variation, and the reconstructed eigenvectors are used as training sets. A probabilistic neural network(PNN) model with fast learning ability and high accuracy was used to build a gyroscope fault diagnosis network, and the trained probabilistic neural network can diagnose the working state of the gyroscope. The EEMD based permutation entropy fault diagnosis, the EEMD based information entropy fault diagnosis, and the EMD based permutation entropy fault diagnosis effect were compared. The research shows that the diagnosis method based on EEMD and permutation entropy algorithm has higher identification ability and higher accuracy for fault state, and can be used as an effective tool for on-line detecting gyroscope faults.
引文
[1] X Chen, B Cui. Efficient modeling of fiber optic gyroscope drift using improved EEMD and extreme learning machine[J]. Singal Processing, 2016,128:1-7.
[2] J Yu, et al. A New Method for Gyroscope Fault Diagnosis Based on CGA RBFNN and Multi-wavelet Entropy[C]. 2013 International Conference on Mechatronic Sciences, Electric Engineering and Computer (MEC) Dec 20-22, 2013.Shenyang,China.IEEE, 345 E 47TH ST, New York, NY 10017 USA:39-43.
[3] Li Liang, Wang Zhenhua, Shen Yi. Fault diagnosis for the intermittent fault in gyroscopes: A data-driven method[C]. Proceedings of the 35th Chinese Control Conference July 27-29,2016,Chengdu,China. IEEE, 345 E 47TH ST, New York, NY 10017 USA: 6639-6643.
[4] Z. Li, et al. Fault detection, identification and reconstruction for gyroscope in satellite based on independent component analysis[J]. Acta Astronautica. 2011,68(7):1015-1023.
[5] 杨杰,等. 基于EEMD-多尺度主元分析的回转支承信号降噪方法研究[J]. 中南大学学报(自然科学版). 2016,47(04):1173-1180.
[6] Yang Jie, et al. Research on the method of signal noise reduction based on EEMD- multiscale principal element analysis[J]. Jornal of Central South University(Science and Technology) . 2016,47(4):1173-1180.
[7] B Cui, X. Chen. Improved hybrid filter for fiber optic gyroscope signal denoising based on EMD and forward linear prediction[J]. Sensors and Actuators A: Physical. 2015,230(Supplement C):150-155.
[8] H N E., et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences. 1998,454(1971):1-41.
[9] Z Wu, N E Huang. Ensemble Empirical Mode Decomposition: A Noise-Assisted Data Analysis Method[J]. Advances in Adaptive Data Analysis. 2009,1(1):1-41.
[10] C Bandt, B Pompe. Permutation Entropy: A Natural Complexity Measure for Time Series[J]. Physical Review Letters. 2002,88(17):174102.
[11] 饶国强,等. 排列熵算法参数的优化确定方法研究[J]. 振动与冲击. 2014,1(36): 188-193.
[12] L Zunino, C W Kulp. Detecting nonlinearity in short and noisy time series using the permutation entropy[J]. Physics Letters A. 2017,381(42):3627-3635.
[13] R Yan, Y Liu, R X Gao. Permutation entropy: A nonlinear statistical measure for status characterization of rotary machines[J]. Mechanical Systems and Signal Processing. 2012,29(Supplement C):474-484.
[14] J A Gutiérrez-Gnecchi, et al. DSP-based arrhythmia classification using wavelet transform and probabilistic neural network[J]. Biomedical Signal Processing and Control. 2017,32(Supplement C):44-56.
[15] J M González-Camacho, et al. Genome-enabledprediction using probabilistic neural network classifiers[J]. BMC Genomics, 2016,17(1): 208.
[16] Y Yu, et al. A novel sensor fault diagnosis method based on Modified EnsembleEmpirical Mode Decomposition and Probabilistic Neural Network[J]. Measurement. 2015,68(Supplement C):328-336.
[17] J Kullaa. Detection, identification, and quantification of sensor fault in a sensor network[J]. Mechanical Systems and Signal Processing. 2013,40(1):208-221.
[18] X Zhang, et al. Feature selection in mixed data: A method using a novel fuzzy rough set-based information entropy[J]. Pattern Recognition. 2016,56(Supplement C):1-15.
[2] J Yu, et al. A New Method for Gyroscope Fault Diagnosis Based on CGA RBFNN and Multi-wavelet Entropy[C]. 2013 International Conference on Mechatronic Sciences, Electric Engineering and Computer (MEC) Dec 20-22, 2013.Shenyang,China.IEEE, 345 E 47TH ST, New York, NY 10017 USA:39-43.
[3] Li Liang, Wang Zhenhua, Shen Yi. Fault diagnosis for the intermittent fault in gyroscopes: A data-driven method[C]. Proceedings of the 35th Chinese Control Conference July 27-29,2016,Chengdu,China. IEEE, 345 E 47TH ST, New York, NY 10017 USA: 6639-6643.
[4] Z. Li, et al. Fault detection, identification and reconstruction for gyroscope in satellite based on independent component analysis[J]. Acta Astronautica. 2011,68(7):1015-1023.
[5] 杨杰,等. 基于EEMD-多尺度主元分析的回转支承信号降噪方法研究[J]. 中南大学学报(自然科学版). 2016,47(04):1173-1180.
[6] Yang Jie, et al. Research on the method of signal noise reduction based on EEMD- multiscale principal element analysis[J]. Jornal of Central South University(Science and Technology) . 2016,47(4):1173-1180.
[7] B Cui, X. Chen. Improved hybrid filter for fiber optic gyroscope signal denoising based on EMD and forward linear prediction[J]. Sensors and Actuators A: Physical. 2015,230(Supplement C):150-155.
[8] H N E., et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences. 1998,454(1971):1-41.
[9] Z Wu, N E Huang. Ensemble Empirical Mode Decomposition: A Noise-Assisted Data Analysis Method[J]. Advances in Adaptive Data Analysis. 2009,1(1):1-41.
[10] C Bandt, B Pompe. Permutation Entropy: A Natural Complexity Measure for Time Series[J]. Physical Review Letters. 2002,88(17):174102.
[11] 饶国强,等. 排列熵算法参数的优化确定方法研究[J]. 振动与冲击. 2014,1(36): 188-193.
[12] L Zunino, C W Kulp. Detecting nonlinearity in short and noisy time series using the permutation entropy[J]. Physics Letters A. 2017,381(42):3627-3635.
[13] R Yan, Y Liu, R X Gao. Permutation entropy: A nonlinear statistical measure for status characterization of rotary machines[J]. Mechanical Systems and Signal Processing. 2012,29(Supplement C):474-484.
[14] J A Gutiérrez-Gnecchi, et al. DSP-based arrhythmia classification using wavelet transform and probabilistic neural network[J]. Biomedical Signal Processing and Control. 2017,32(Supplement C):44-56.
[15] J M González-Camacho, et al. Genome-enabledprediction using probabilistic neural network classifiers[J]. BMC Genomics, 2016,17(1): 208.
[16] Y Yu, et al. A novel sensor fault diagnosis method based on Modified EnsembleEmpirical Mode Decomposition and Probabilistic Neural Network[J]. Measurement. 2015,68(Supplement C):328-336.
[17] J Kullaa. Detection, identification, and quantification of sensor fault in a sensor network[J]. Mechanical Systems and Signal Processing. 2013,40(1):208-221.
[18] X Zhang, et al. Feature selection in mixed data: A method using a novel fuzzy rough set-based information entropy[J]. Pattern Recognition. 2016,56(Supplement C):1-15.