优化BP神经网络的光纤陀螺温度漂移建模与补偿
|
摘要
光纤陀螺(FOG)温度漂移误差是影响其输出精度的主要误差源之一。针对基于传统BP神经网络FOG温度误差补偿方案适用性较差的问题,提出了优化预测数据的BP神经网络补偿算法,利用最优线性平滑技术以及滑动平均技术对神经网络待补偿数据进行预处理,可以有效减小FOG输出白噪声对温度漂移网络模型补偿精度的干扰,优化神经网络模型的补偿效果。使用FOG温度漂移实测数据对所提出的优化算法进行验证,结果表明利用本文提出的两种建模及补偿方案进行补偿后的FOG温度漂移数据标准差相比传统BP神经网络补偿方法减少50%以上。
In view of the poor applicability of traditional BP neural network's compensation algorithm, a new BP neural network with optimized prediction data was proposed by using an optimal liner smoothing technique and a moving average technique to pre-process the data for neural network compensation, which effectively reduces the white noise disturbance's influences on the compensation results of neural network model. The verification of the proposed optimization algorithm is made by using FOG temperature drift data, which shows that the standard drift of FOG temperature drift data is reduced to below 50% by using the proposed two modeling and compensation schemes, compared with traditional BP neural network compensation method.
In view of the poor applicability of traditional BP neural network's compensation algorithm, a new BP neural network with optimized prediction data was proposed by using an optimal liner smoothing technique and a moving average technique to pre-process the data for neural network compensation, which effectively reduces the white noise disturbance's influences on the compensation results of neural network model. The verification of the proposed optimization algorithm is made by using FOG temperature drift data, which shows that the standard drift of FOG temperature drift data is reduced to below 50% by using the proposed two modeling and compensation schemes, compared with traditional BP neural network compensation method.
引文
[1]杨孟兴,陈俊杰.光纤陀螺静态温度特性的分析及实验研究[J].中国惯性技术学报,2011,18(6):751-755.Yang Meng-xing,Chen Jun-jie.Analysis and experiment on static temperature characteristic of FOG[J].Journal of Chinese Inertial Technology,2011,18(6):751-755.
[2]韩冰,林玉荣,邓正隆.光纤陀螺温度漂移误差的建模与补偿综述[J].中国惯性技术学报,2009,17(2):218-224.Han Bing,Lin Yu-rong,Deng Zheng-long.Overview on modeling and compensation of FOG temperature drift[J].Journal of Chinese Inertial Technology,2009,17(2):218-224.
[3]张岩岫,段红,祝建成.神经网络在光纤陀螺温度漂移中的应用[J].陕西科技大学学报,2008,26(5):95-98.Zhang Yan-xiu,Duan Hong,Zhu Jian-cheng.Application of Neutral Network in the FOG Temperature Compensation[J].Journal of Shanxi University of Science&Technology,2008,26(5):95-98.
[4]周琪,秦永元,赵长山.光纤陀螺温度漂移误差的模糊补偿方案研究[J].传感技术学报,2010,23(7):926-930.Zhou QI,Qin Yong-yuan,Zhao Chang-shan.Research on fuzzy compensation method of temperature drift for fiber optical gyro[J].Chinese Journal of Sensors and Actuators,2010,23(7).
[5]Narasimhappa M,Sabat S L,Rangababu Peesapati R,et al.An innovation based random weighting estimation mechanism for denoising fiber optic gyro drift signal[J].Optik-International Journal for Light and Electron Optics,2014,125(3):1192-1198.
[6]Li Chuan-sheng,Zhang Chun-xi,Wang Xia-xiao,et al.Analysis and compensation of ratio temperature error for sagnac fiber-optic current transformer[J].Electric Power Automation Equipment,2012,32(11):102-106,114.
[7]冯卡力,李安,覃方君.基于多模型分段拟合的光纤陀螺温度误差补偿方法[J].中国惯性技术学报,2014,22(6):825-828.Feng Ka-li,Li An,Qin Fang-jun.Temperature error compensation method for FOG based on multi-model piecewise fitting[J].Journal of Chinese Inertial Technology,2014,22(6):825-828.
[8]Jadav K,Panchal M.Optimizing weights of artificial neural networks using genetic algorithms[J].International Journal of Advanced Research in Computer Science and Electronics Engineering,2012,1(10):47-51.
[9]Donate J P,Cortez P,Sánchez G,et al.Time series forecasting using a weighted cross-validation evolutionary artificial neural network ensemble[J].Neuro Computing,2013,109:27-32.
[10]Olyaee S,Ebrahimpour R,Hamedi S.Modeling and compensation of periodic nonlinearity in two-mode interferometer using neural networks[J].IETE Journal of Research,2010,56(2):102-110.
[11]Corporation H P.Analysis and optimization of dynamic measurement precision of fiber optic gyroscope[J].Mathematical Problems in Engineering,2013(2):142-148.
[12]Xu X,Zhang C,Pan X.Study of reflection error in closed-loop polarization-maintained interferometric fiber optic gyroscope[J].Optik-International Journal for Light and Electron Optics,2010,121(13):1170-1175.
[13]Lli Jie-yu,Yang Jian-ye,Wang Li-xin,et al.Error autocompensation of FOG-based rotating inertial measurement unit for ballistic missile[J].Opto-Electronic Engineering,2012,39(6):41-46.
[2]韩冰,林玉荣,邓正隆.光纤陀螺温度漂移误差的建模与补偿综述[J].中国惯性技术学报,2009,17(2):218-224.Han Bing,Lin Yu-rong,Deng Zheng-long.Overview on modeling and compensation of FOG temperature drift[J].Journal of Chinese Inertial Technology,2009,17(2):218-224.
[3]张岩岫,段红,祝建成.神经网络在光纤陀螺温度漂移中的应用[J].陕西科技大学学报,2008,26(5):95-98.Zhang Yan-xiu,Duan Hong,Zhu Jian-cheng.Application of Neutral Network in the FOG Temperature Compensation[J].Journal of Shanxi University of Science&Technology,2008,26(5):95-98.
[4]周琪,秦永元,赵长山.光纤陀螺温度漂移误差的模糊补偿方案研究[J].传感技术学报,2010,23(7):926-930.Zhou QI,Qin Yong-yuan,Zhao Chang-shan.Research on fuzzy compensation method of temperature drift for fiber optical gyro[J].Chinese Journal of Sensors and Actuators,2010,23(7).
[5]Narasimhappa M,Sabat S L,Rangababu Peesapati R,et al.An innovation based random weighting estimation mechanism for denoising fiber optic gyro drift signal[J].Optik-International Journal for Light and Electron Optics,2014,125(3):1192-1198.
[6]Li Chuan-sheng,Zhang Chun-xi,Wang Xia-xiao,et al.Analysis and compensation of ratio temperature error for sagnac fiber-optic current transformer[J].Electric Power Automation Equipment,2012,32(11):102-106,114.
[7]冯卡力,李安,覃方君.基于多模型分段拟合的光纤陀螺温度误差补偿方法[J].中国惯性技术学报,2014,22(6):825-828.Feng Ka-li,Li An,Qin Fang-jun.Temperature error compensation method for FOG based on multi-model piecewise fitting[J].Journal of Chinese Inertial Technology,2014,22(6):825-828.
[8]Jadav K,Panchal M.Optimizing weights of artificial neural networks using genetic algorithms[J].International Journal of Advanced Research in Computer Science and Electronics Engineering,2012,1(10):47-51.
[9]Donate J P,Cortez P,Sánchez G,et al.Time series forecasting using a weighted cross-validation evolutionary artificial neural network ensemble[J].Neuro Computing,2013,109:27-32.
[10]Olyaee S,Ebrahimpour R,Hamedi S.Modeling and compensation of periodic nonlinearity in two-mode interferometer using neural networks[J].IETE Journal of Research,2010,56(2):102-110.
[11]Corporation H P.Analysis and optimization of dynamic measurement precision of fiber optic gyroscope[J].Mathematical Problems in Engineering,2013(2):142-148.
[12]Xu X,Zhang C,Pan X.Study of reflection error in closed-loop polarization-maintained interferometric fiber optic gyroscope[J].Optik-International Journal for Light and Electron Optics,2010,121(13):1170-1175.
[13]Lli Jie-yu,Yang Jian-ye,Wang Li-xin,et al.Error autocompensation of FOG-based rotating inertial measurement unit for ballistic missile[J].Opto-Electronic Engineering,2012,39(6):41-46.