双自适应BP算法在智能手环中的应用
|
摘要
智能手环体征信息检测的准确性与及时性直接影响到它的使用和推广。通过多传感器数据融合技术对人体体征信息进行检测是主流趋势。针对标准BP神经网络算法存在的收敛速度慢、容易陷入局部极值的缺陷,增加了自适应动量法和自适应学习率调整法以改进传统BP算法,给出了改进后的BP网络模型及参数设计,并应用于智能手环体征检测工作中。仿真实验证明,改进的算法能够使BP网络快速收敛,且提高了手环系统报警的准确率,减少了误判的发生。
The accuracy and timeliness of smart bracelet sign detection directly affect its use and promotion.For human body sign information detection,the multi-sensor data fusion technology is needed.In view of the defects of slow convergence rate and easy to fall into local extremum,the adaptive momentum method and adaptive learning rate adjustment method are added to improve the traditional BP algorithm.The improved BP network model and parameter design are given,and it is applied to the detection of intelligent hand loop(smart BP).The simulation experiment results show that the performance of the improved algorithm is obviously superior to the traditional BP algorithm.It can make the BP network converge quickly,and improve the accuracy of the alarm system of the hand loop system and reduce the occurrence of miscarriage of justice.
The accuracy and timeliness of smart bracelet sign detection directly affect its use and promotion.For human body sign information detection,the multi-sensor data fusion technology is needed.In view of the defects of slow convergence rate and easy to fall into local extremum,the adaptive momentum method and adaptive learning rate adjustment method are added to improve the traditional BP algorithm.The improved BP network model and parameter design are given,and it is applied to the detection of intelligent hand loop(smart BP).The simulation experiment results show that the performance of the improved algorithm is obviously superior to the traditional BP algorithm.It can make the BP network converge quickly,and improve the accuracy of the alarm system of the hand loop system and reduce the occurrence of miscarriage of justice.
引文
[1]王子洪.基于WSN穿戴式系统的研究现状与展望[J].中国医疗设备,2012,27(2):62-65.
[2]黄自鑫,薛勇,刘旭.可穿戴无线体域网中异质传感器数据融合算法研究[J].自动化与控制,2015(5):145-147.
[3]石九龙.基于老人监护用途的可穿戴设备的设计与实现[D].武汉:华中师范大学,2015.
[4]王犇.多传感器信息融合及其在可穿戴计算机上的应用[D].哈尔滨:哈尔滨工业大学,2008.
[5]王海洋.基于BP神经网络的数据融合算法及其在分布式智能养老系统中应用[D].沈阳:沈阳大学,2013.
[6]王静.基于数据融合的分布式智能养老系统的应用研究[D].沈阳:沈阳大学,2014.
[7]廖武.体域网中多传感器数据融合算法的研究[D].长春:吉林大学,2016.
[8]熊峰,龙红叶,胡小梅,等.动量因子BP神经网络算法在设备故障预测中的应用[J].制造业自动化,2011,33(12):13-16.
[9]邓子云.基于有动量和自适应lr梯度下降法BP神经网络的城市用电量预测技术[J].江苏科技信息,2018(3):32-34.
[10]王文成.BP神经网络中自适应学习率的研究[J].计算机科学,1995,22(4):48-50.
[11]李康,刘辉,罗彬.引入动量因子的双自适应自然梯度算法[J].湖南师范大学自然科学学报,2015,38(1):54-57.
[12]付光杰,胡明哲.基于改进型学习率自适应BP神经网络的风力发电变桨距控制技术[J].自动化与仪器仪表,2016(6):8-10.
[13]冯国良,秦晓明.二次自适应调整学习参数的改进型BP算法研究[J].硅谷,2009(1):118-131.
[14]CICHOCHI A,UNBEHAUEN R.Robust neural networks with on-line learning for blind identification and blind separation of sources[J].IEEE Trans on Circuit and systems:Fundamental Theory and Applications,1996,43(11):894-906.
[15]李小寒,尚少梅.基础护理学[M].5版.北京:人民卫生出版社,2014.
[2]黄自鑫,薛勇,刘旭.可穿戴无线体域网中异质传感器数据融合算法研究[J].自动化与控制,2015(5):145-147.
[3]石九龙.基于老人监护用途的可穿戴设备的设计与实现[D].武汉:华中师范大学,2015.
[4]王犇.多传感器信息融合及其在可穿戴计算机上的应用[D].哈尔滨:哈尔滨工业大学,2008.
[5]王海洋.基于BP神经网络的数据融合算法及其在分布式智能养老系统中应用[D].沈阳:沈阳大学,2013.
[6]王静.基于数据融合的分布式智能养老系统的应用研究[D].沈阳:沈阳大学,2014.
[7]廖武.体域网中多传感器数据融合算法的研究[D].长春:吉林大学,2016.
[8]熊峰,龙红叶,胡小梅,等.动量因子BP神经网络算法在设备故障预测中的应用[J].制造业自动化,2011,33(12):13-16.
[9]邓子云.基于有动量和自适应lr梯度下降法BP神经网络的城市用电量预测技术[J].江苏科技信息,2018(3):32-34.
[10]王文成.BP神经网络中自适应学习率的研究[J].计算机科学,1995,22(4):48-50.
[11]李康,刘辉,罗彬.引入动量因子的双自适应自然梯度算法[J].湖南师范大学自然科学学报,2015,38(1):54-57.
[12]付光杰,胡明哲.基于改进型学习率自适应BP神经网络的风力发电变桨距控制技术[J].自动化与仪器仪表,2016(6):8-10.
[13]冯国良,秦晓明.二次自适应调整学习参数的改进型BP算法研究[J].硅谷,2009(1):118-131.
[14]CICHOCHI A,UNBEHAUEN R.Robust neural networks with on-line learning for blind identification and blind separation of sources[J].IEEE Trans on Circuit and systems:Fundamental Theory and Applications,1996,43(11):894-906.
[15]李小寒,尚少梅.基础护理学[M].5版.北京:人民卫生出版社,2014.