【作者】 宋春生；

【导师】 周祖德；

【作者基本信息】 武汉理工大学 ， 机械制造及其自动化， 2011， 博士

【摘要】 舰船声隐身性能的好坏对我国国防安全具有极其重要的意义。隔振技术能够有效的抑制结构振动的能量传递,成为潜艇减振降噪的关键技术。隔振分为主动隔振和被动隔振两种形式。被动隔振由其自身所限,对低频干扰隔振能力差,特别是对系统谐振频率附近的干扰信号还有放大作用。主动隔振可以根据设定的控制规律动态地调整系统的支承特性参数,满足被动隔振无法实现的低频及谐振频率附近的隔振降噪的需要。因此,充分利用主被动隔振技术的优势,研究主被动结合的隔振技术成为当今研究的热点。而采用磁悬浮支承技术的磁悬浮隔振器,具有无接触、寿命长、响应快和电磁力及支承参数(刚度,阻尼等)随控制参数的变化可控可调的特点,是一种较理想的主动隔振器。将磁悬浮隔振器应用于传统的被动隔振系统(如单层、双层及浮筏系统)中,能有效弥补被动隔振系统低频及谐振区隔振性能的不足,具有重要的研究意义。首先,采用功率流作为隔振评价指标,从两个自由度和多个自由度隔振系统出发,分别建立了两类隔振系统模型,分析了被动隔振系统参数对隔振性能的影响。研究表明在其它结构参数一定的情况下,隔振器的刚度和阻尼参数是影响隔振性能的主要参数。而磁悬浮隔振器能通过调节电磁力的大小,使其支承刚度和阻尼在线可控可调,因此磁悬浮隔振器在隔振系统应用的优势明显。建立了磁悬浮隔振器的数学模型,选择适当的参数,设计了磁悬浮隔振器。根据实际工况,考虑衔铁和E型磁铁的铁磁材料的磁饱和、漏磁、磁场耦合等因素,建立了磁悬浮隔振系统的三维静态电磁场有限元模型,计算了电磁场分布。在此基础上,选择适当的测点,采用三通道高斯计对磁悬浮隔振器的磁场分布进行测量,实验结果与有限元分析结果基本吻合,验证了有限元模型的准确性。对磁悬浮隔振器的静动电磁力进行实际测量,静态电磁力利用最小二乘法进行拟合,进而对理论电磁力公式参数进行修正,得到了实际的磁悬浮隔振器静态电磁力、位移与电流之间的关系；针对动态电磁力非线性及滞后的特征,采用基于改进的遗传算法的BP神经网络对磁悬浮隔振器动态电磁力进行模型辨识,建立了动态电磁力、位移与电流之间的关系,为主动隔振控制提供可靠依据。将磁悬浮隔振器应用于被动隔振系统,采用刚体动力学、动态子结构及有限元等方法,建立了磁悬浮单层、双层以及浮筏隔振系统主动系统动力学方程及对应的状态方程,并利用ANSYS软件建立起柔性浮筏有限元模型,进行了仿真分析。选用最小输出力作为磁悬浮隔振系统隔振控制指标,建立了控制模型,推导出了性能指标函数表达式,考虑磁悬浮隔振器的实际能力,对此性能指标函数进行修正,得到修正的性能指标函数。在此基础上,推导出最优反馈矩阵,最后对磁悬浮单层系统与浮筏系统分别进行仿真分析,证明了控制策略的有效性。针对磁悬浮隔振器非线性的特点,结合提出的控制策略,采用模糊控制方法,建立了磁悬浮主动隔振模糊控制器,进行仿真分析。仿真结果表明,隔振效果优于被动系统和PID控制隔振系统。搭建了磁悬浮隔振实验系统,对比研究主被动隔振系统隔振效果,实验结果与仿真结果基本吻合。考虑到筏架柔性因素,为了建立更为准确柔性浮筏的动力学模型,获取柔性筏架的实时动力特性非常重要,、因此本文最后提出了一种基于光纤布喇格光栅分布式传感器柔性筏架测量技术,能准确实时的获取简化的柔性筏架模型的模态及应变特性。更多还原

【Abstract】 It has the extremely vital significance to the national defense security to improve the performance of the naval vessels acoustic stealthy. Vibration isolation technology widely used to reduce vibration transmission in many different engineering applications has become a key technique of vibration and noise reduction. It is traditionally classified into two types:passive, active. Although the passive systems offer simple and reliable means to protect mechanical systems from their respective environments, they have inherent performance limitations including their limited controllable frequency range and their unchangeable shapes of transmissibility. Active vibration isolation systems, on the other hand, can provide’enhanced vibration isolation performance through a feedback system. However, active vibration isolation systems are not cost-effective. Therefore, passive and active combined technology is regarded as a hotspot in current researches, due to performance improvement and simplicity of implementation. Magnetic Suspension Vibration Isolation(MSVI) technology is an excellent active isolation technology, which has some useful characteristics, such as non-contact, high response frequency, high reliability and long life-span, especially, the damping and stiffness adjusted by changing controller’s parameters on-line. Therefore, it is significance that the technology is applied in the passive vibration isolation system in this research to improve the isolation performance.A two degree of freedom system and an multiple degree of freedom system models are established to analyze effect of system parameters on vibration isolation performance using power flow as evaluation criterion. Results show that with all other parameters definition, the stiffness and damping are the key parameters of influencing the performance of vibration isolation. And the MSVI technology has obvious advantages to be applied to change stiffness and damping parameter.The principle of magnetic suspension isolation technology is introduced. The math modeling of the Magnetic Suspension Isolator(MSI) is established. The MSI is designed according to practical parameters. For the work status of the MSI, static electromagnetic field distribution is calculated through 3D FEM with consideration of nonlinearities of armature and E-shape magnet magnetic, including material’s nonlinear, magnetic saturation and leakage, magnetic coupling, and etc. On the basis of the theoretical analysis, the electromagnetic field distribution is measure using a three channel Guass meter to ascertain the accuracy of the FEM model at some specific spots. Then, the static and dynamic electromagnetic force of the MSI are measured by experiment. Through the experimental measurements and fitting a mathematical expression by least square method, the actual relation expression of static electromagnetic force-current-gap is obtained for accuracy control. In view of highly non-linear and hysteresis behavior of dynamic electromagnetic force, Artificial Neural Networks(ANN) is applied to identify the relationship of dynamic electromagnetic force. To achieve higher accuracy for ANN, an modified genetic algorithm is applied to train the neural networks. Results clearly show the MGA based neural networks is found better performance and higher accurate for MSI model comparing with conventional BP algorithm.The designed MSI is applied to the passive vibration isolation system. The dynamical equations and state equations of the single-layer, double-layer and (flexible) floating raft system are built, using the methods of rigid body dynamics, dynamic sub-structure, Finite Element methods.A control mechanism and a cost function of active vibration isolation based on the minimization of the weighted sum of squared output forces are proposed. Considering the practical capacity of the MSI, the value function is revised. The revised function and the optimal feedback matrix are deduced. A control model based on the control mechanism is established. The results of simulation show validity of the strategy. For the nonlinear of the MSI, according to the control strategy of the former, an output feedback fuzzy control algorithm is proposed to the active isolation system. The fuzzy control model are simulated using Matlab fuzzy logical toolbox. The results of simulation show that the fuzzy controller possesses better isolation performance than passive system and PID control system.In order to test the performance of the active system and control algorithm, an experimental platform is carried out. The experiment results show that the active isolation performance improve highly at around the resonant frequencies compared to the passive system. The experimental results are found to be in good agreement with the simulated results.At last, to solve the strain multi-position measurement of raft frame, the Fiber Bragg Grating(FBG) is introduced. According to the simplicity principles of raft, a platform is built to measure the stress and modal of simple raft using FBG sensors. The experimental results are found to be in good agreement with the FEA results.更多还原