基于QPSO-Volterra的齿轮裂纹故障特征提取
|
摘要
鉴于目前主流齿轮裂纹故障检测方法所存在的局限性(即仅利用系统响应作为研究对象,很少考虑输入对于故障特征提取的作用),并考虑到其作为一种典型非线性系统所蕴含的动态特性,将Volterra级数理论应用于不同状态齿轮啮合传动系统,以充分发挥Volterra级数能够综合利用系统输入、输出数据进行系统非线性特性描述的优势;同时考虑到QPSO算法较高的全局搜索能力,采用该算法对齿轮啮合传动系统Volterra模型进行了时域核辨识。仿真实验结果表明,高阶时域核对于齿轮裂纹故障所引起的系统非线性特性变化非常敏感,可以有效地表征并区分出不同状态下齿轮啮合传动系统的非线性动态特性,达到了预期目的。
In view of the limitation of current mainstream methods of gear crack fault detection(that is only using the system response as research object, and seldom considering the input effect on fault feature extraction), and taking into account its dynamic characteristics as a kind of typical nonlinear system, by applying Volterra series theory into different states of gear meshing transmission system, for fully taking the advantage of Volterra series, which can use both the input and output data of system to describe the nonlinear characteristics of system. Meanwhile, using the high global search capability of QPSO algorithm, QPSO algorithm is used to identify the time domain kernel of gear meshing transmission system's Volterra model(referred to as GIRF). The simulation results show that the high-order GIRFs are very sensitive to the nonlinear characteristics of the system caused by the gear crack failure, and can effectively characterize and distinguish the nonlinear dynamic characteristics of the gear meshing transmission system under different conditions. The simulation results achieve the expected purpose.
In view of the limitation of current mainstream methods of gear crack fault detection(that is only using the system response as research object, and seldom considering the input effect on fault feature extraction), and taking into account its dynamic characteristics as a kind of typical nonlinear system, by applying Volterra series theory into different states of gear meshing transmission system, for fully taking the advantage of Volterra series, which can use both the input and output data of system to describe the nonlinear characteristics of system. Meanwhile, using the high global search capability of QPSO algorithm, QPSO algorithm is used to identify the time domain kernel of gear meshing transmission system's Volterra model(referred to as GIRF). The simulation results show that the high-order GIRFs are very sensitive to the nonlinear characteristics of the system caused by the gear crack failure, and can effectively characterize and distinguish the nonlinear dynamic characteristics of the gear meshing transmission system under different conditions. The simulation results achieve the expected purpose.
引文
[1]张德祥,汪萍,吴小培.Hilbert-小波变换的齿轮箱故障检测[J].计算机应用研究,2011,28(11):4236-4239.
[2]薛灿灿.基于定子电流双树复小波包分析的机车齿轮故障诊断[D].北京:北京交通大学,2017:11-70.
[3]师宁宁.基于Alpha稳定分布参数的齿轮故障建模及检测[D].哈尔滨:哈尔滨工业大学,2013:18-48.
[4]孙守保,郭瑜,伍星.基于共振区幅值-相位解调技术的早期齿轮齿根裂纹故障检测[J].机械强度,2017,39(1):7-13.
[5]杨定新,胡茑庆,杨银刚,等.随机共振技术在齿轮箱故障检测中的应用[J].振动工程学报,2004,17(2):201-204.
[6]焦李成.非线性系统故障诊断的伏尔泰拉泛函理论[J].西安交通大学学报,1988,22(2):79-85.
[7]欧文,韩崇昭,王文正.Volterra泛函级数在非线性系统辨识中的应用[J].控制与决策,2002,17(2):239-242.
[8]李志农,蒋静,陈金刚,等.基于量子粒子群优化的Volterra核辨识算法研究[J].振动与冲击,2013,32(3):60-63.
[9]魏瑞轩,韩崇昭,张优云,等.非线性系统故障诊断的Volterra模型方法[J].系统工程与电子技术,2004,26(11):1736-1739.
[10]BEDROSIAN E,RICE S.O.The output properties of Volterra systems(nonlinear systems with memory)driven by harmonic and Gaussian inputs[J].Proceedings of the IEEE,1971,59(12):1688-1707.
[11]MATHEWS V J.Adaptive Volterra filters using orthogonal structures[J].IEEE Signal Processing Letters,1996,3(12):307-309.
[12]袁海英,王铁流,陈光.基于Volterra频域核和神经网络的非线性模拟电路故障诊断法[J].仪器仪表学报,2007,28(5):807-811.
[13]夏鑫,周建中,朱文龙,等.基于Volterra模型盲辨识的水轮机转轮状态识别[J].中国电机工程学报,2014,34(20):3392-3396.
[14]唐浩,屈梁生,温广瑞.基于Volterra级数的转子故障诊断研究[J].中国机械工程,2009,20(4):447-454.
[15]王海涛,张霄,史丽晨,等.Volterra核函数法在轴承滚珠磨损中的特征提取及应用[J].应用数学和力学,2017,38(6):633-642.
[16]祝志文,石亚光,李健朋.基于Volterra理论的扁平箱梁气动力非线性特性研究[J].中国公路学报,2018,31(1):74-81.
[17]袁杰.基于Volterra级数的射频功放数字预失真技术研究[D].郑州:郑州大学,2016:34-44.
[18]欧阳文.基于二阶Volterra级数滤波器的海杂波建模研究[J].电光与控制,2016,23(1):29-32.
[19]王西.齿轮故障的动力学建模与轮齿裂纹刚度计算方法研究[D].重庆:重庆大学,2012:27-36.
[20]WU S Y,ZUO M J,PAREY A.Simulation of spur gear dynamics and estimation of fault growth[J].Journal of Sound and Vibration,2008,317(3/5):608-624.
[21]孙俊.量子行为粒子群优化算法研究[D].无锡:江南大学,2009:12-23.
[22]TIAN X H.Dynamic simulation for system response of gearbox including localized gearfaults[D].Edmonton:University of Albert,2004:56.
[2]薛灿灿.基于定子电流双树复小波包分析的机车齿轮故障诊断[D].北京:北京交通大学,2017:11-70.
[3]师宁宁.基于Alpha稳定分布参数的齿轮故障建模及检测[D].哈尔滨:哈尔滨工业大学,2013:18-48.
[4]孙守保,郭瑜,伍星.基于共振区幅值-相位解调技术的早期齿轮齿根裂纹故障检测[J].机械强度,2017,39(1):7-13.
[5]杨定新,胡茑庆,杨银刚,等.随机共振技术在齿轮箱故障检测中的应用[J].振动工程学报,2004,17(2):201-204.
[6]焦李成.非线性系统故障诊断的伏尔泰拉泛函理论[J].西安交通大学学报,1988,22(2):79-85.
[7]欧文,韩崇昭,王文正.Volterra泛函级数在非线性系统辨识中的应用[J].控制与决策,2002,17(2):239-242.
[8]李志农,蒋静,陈金刚,等.基于量子粒子群优化的Volterra核辨识算法研究[J].振动与冲击,2013,32(3):60-63.
[9]魏瑞轩,韩崇昭,张优云,等.非线性系统故障诊断的Volterra模型方法[J].系统工程与电子技术,2004,26(11):1736-1739.
[10]BEDROSIAN E,RICE S.O.The output properties of Volterra systems(nonlinear systems with memory)driven by harmonic and Gaussian inputs[J].Proceedings of the IEEE,1971,59(12):1688-1707.
[11]MATHEWS V J.Adaptive Volterra filters using orthogonal structures[J].IEEE Signal Processing Letters,1996,3(12):307-309.
[12]袁海英,王铁流,陈光.基于Volterra频域核和神经网络的非线性模拟电路故障诊断法[J].仪器仪表学报,2007,28(5):807-811.
[13]夏鑫,周建中,朱文龙,等.基于Volterra模型盲辨识的水轮机转轮状态识别[J].中国电机工程学报,2014,34(20):3392-3396.
[14]唐浩,屈梁生,温广瑞.基于Volterra级数的转子故障诊断研究[J].中国机械工程,2009,20(4):447-454.
[15]王海涛,张霄,史丽晨,等.Volterra核函数法在轴承滚珠磨损中的特征提取及应用[J].应用数学和力学,2017,38(6):633-642.
[16]祝志文,石亚光,李健朋.基于Volterra理论的扁平箱梁气动力非线性特性研究[J].中国公路学报,2018,31(1):74-81.
[17]袁杰.基于Volterra级数的射频功放数字预失真技术研究[D].郑州:郑州大学,2016:34-44.
[18]欧阳文.基于二阶Volterra级数滤波器的海杂波建模研究[J].电光与控制,2016,23(1):29-32.
[19]王西.齿轮故障的动力学建模与轮齿裂纹刚度计算方法研究[D].重庆:重庆大学,2012:27-36.
[20]WU S Y,ZUO M J,PAREY A.Simulation of spur gear dynamics and estimation of fault growth[J].Journal of Sound and Vibration,2008,317(3/5):608-624.
[21]孙俊.量子行为粒子群优化算法研究[D].无锡:江南大学,2009:12-23.
[22]TIAN X H.Dynamic simulation for system response of gearbox including localized gearfaults[D].Edmonton:University of Albert,2004:56.