基于次级通道在线辨识的齿轮传动系统振动主动控制研究

发布时间：2018-09-10 08:39

【摘要】：齿轮作为一种通用件，是各种机械装备系统中的重要传动装置，在机械、交通、化工、航空、航天、船舶等行业的设备中起着非常关键的作用。实际应用中的齿轮不可避免地存在着制造、安装误差，在载荷的作用下轮齿将发生变形。这些误差和变形破坏了齿轮传动的啮合关系，使齿轮啮合时的位置相对于其理论位置发生偏离，从而使瞬时传动比发生变化，造成齿与齿之间的碰撞和冲击，形成了齿轮啮合的内部激励，从而产生振动和噪声。本文结合压电智能材料的应用，将主动控制技术用于齿轮啮合传动系统振动控制，基于轴横向振动主动控制思想，提出了一种齿轮箱内的主动控制结构，使主动控制力以更直接的方式控制误差激励引起的振动响应。利用多组压电片在力学上串联、电学上并联的设计思想，构成压电堆作动器，基于FxLMS自适应滤波算法，通过作动器输出的位移控制齿轮轴的弯曲振动，进而达到衰减振动和噪声的目的。具体工作包括以下几个方面： 1)采用集中质量法建立了直齿轮三自由度振动数学模型，并基于此模型对齿轮系统动态响应特性进行分析，为主动控制奠定了基础。分析了激励频率、负载、阻尼等对齿轮振动状态和啮合形式的影响；对比了线性模型和非线性模型，时变刚度模型和时不变刚度模型用于动力学分析时的区别。 2)基于控制轴横向振动的概念，提出一种齿轮传动系统振动主动控制结构。采用有限元的方法，建立了主动控制结构的数学模型，为仿真研究奠定基础。阐述了FxLMS自适应滤波算法的基本原理，并用FxLMS算法进行了主动控制仿真研究;分析了次级通道误差和参考信号对FxLMS算法收敛性能和稳定性的影响。 3)分析了对次级通道进行在线辨识的原因，讨论了附加随机信号进行在线辨识的几种算法，分析了其性能特点，并进行了主动控制仿真验证。将DLMS算法应用到振动主动控制中，采用自适应的方法对次级通道进行延迟估计，并进行了仿真验证。 4)搭建了齿轮传动系统振动主动控制实验平台，开展了振动主动控制研究。首先在离线条件下研究了次级通道模型，，并对采用自适应陷波器获取参考信号的方法进行了实验验证。然后设计相应控制器，对在线辨识的自适应滤波算法进行了振动主动控制实验研究，验证了主动控制算法和结构的有效性。
[Abstract]:As a universal part, gear is an important transmission device in various mechanical equipment systems, and plays a very important role in machinery, transportation, chemical industry, aviation, aerospace, ship and other industries. In practical application, there are inevitable manufacturing and installation errors, and the gear teeth will deform under the action of load. These errors and deformation destroy the meshing relation of gear transmission, make the position of gear meshing deviate from its theoretical position, and make the instantaneous transmission ratio change, resulting in the collision and impact between teeth. The internal excitation of gear meshing is formed, resulting in vibration and noise. Combined with the application of piezoelectric intelligent material, the active control technology is applied to the vibration control of gear meshing transmission system. Based on the idea of active control of axial transverse vibration, a kind of active control structure in gear box is proposed. The vibration response caused by error excitation is controlled more directly by active control force. Based on the design idea of series piezoelectric plates in mechanics and parallel in electricity, a piezoelectric pile actuator is constructed. Based on FxLMS adaptive filtering algorithm, the bending vibration of gear shaft is controlled by the displacement of actuator output. Then the aim of attenuating vibration and noise is achieved. The specific work includes the following aspects: 1) the mathematical model of three degrees of freedom vibration of spur gear is established by means of concentrated mass method, and the dynamic response characteristics of gear system are analyzed based on this model, which lays a foundation for active control. The effects of excitation frequency, load and damping on the vibration state and meshing form of gear are analyzed, and the linear model and nonlinear model are compared. The difference between time-varying stiffness model and time-invariant stiffness model in dynamic analysis. 2) based on the concept of lateral vibration of control shaft, a new active vibration control structure for gear transmission system is proposed. The mathematical model of active control structure is established by finite element method, which lays a foundation for simulation research. The basic principle of FxLMS adaptive filtering algorithm is expounded, and the active control simulation is carried out with FxLMS algorithm. The influence of secondary channel error and reference signal on the convergence performance and stability of FxLMS algorithm is analyzed. 3) the reason for on-line identification of secondary channel is analyzed, and several algorithms for on-line identification of additional random signals are discussed. The performance characteristics are analyzed, and the active control simulation is carried out. The DLMS algorithm is applied to the active vibration control, and the adaptive method is used to estimate the delay of the secondary channel. 4) an experimental platform for active vibration control of the gear transmission system is built. Active vibration control is studied. First, the secondary channel model is studied under off-line condition, and the method of obtaining reference signal by adaptive notch is verified experimentally. Then the corresponding controller is designed and the active vibration control experiment is carried out to verify the validity of the active control algorithm and structure.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TH132.41;TB535

【相似文献】