【作者】 徐震；

【导师】 邓华；

【作者基本信息】 中南大学 ， 机械工程， 2013， 博士

【摘要】 重载锻造操作机是一种与自由锻压机联合实现锻件锻造的大型生产设备。重载锻造操作机在夹持重达上百吨,直径有4m锻件的情况下,所需的驱动力矩也达到上百吨米,单个驱动系统很难达到快速启停定位的要求,所以必须采用并联驱动方式才能解决。然而,并联驱动系统动态特性的差异、间隙和摩擦等因素引起驱动作用力的不均衡,造成驱动机构偏载、力流畸变等不确定力学行为,降低了驱动机构的使用寿命甚至功能失效。因此,必须对锻造操作机夹持系统并联驱动均载性进行研究。由于锻造操作机是与自由锻压机配合实现自动锻造操作,锻造操作机夹持系统必须进行频繁快速启停的非连续旋转运动,此时驱动机构中的齿轮传动是在一种矩形类脉冲间歇驱动方式下进行,而齿轮间隙会引起齿轮传动的冲击现象更加明显,并且系统中间环节的变化比如齿轮传动的作用力变化是无法直接测量得到,所以需要一种有效求解齿轮接触／冲击动力学方程的计算方法对齿轮传动在间歇冲击驱动下的动力学行为进行研究。另外,重载锻造操作机夹持系统转动惯量大、液压系统元件的强非线性、齿轮的间隙和摩擦等非线性和不确定性,对锻造操作机夹持系统并联驱动在时变重载工况下的均载性和快速启停定位造成不同程度的影响,所以必须对锻造操作机夹持系统并联驱动进行均载分析与均载控制研究。本文针对上述问题进行了以下主要研究工作：针对锻造操作机在频繁启停、间歇驱动下非连续运动的齿轮接触／冲击问题,提出了基于Newmark积分的动量迭代运算方法。该方法将力的二阶运动方程变为动量的一阶运动方程,便于求解计算,同时消除了系统频繁启停造成的非连续运动影响。通过与齿轮非连续运动的三维有限元动力学仿真和实验结果比较,验证了该方法的合理性且保证了计算精度和效率。为了研究重载锻造操作机夹持系统并联驱动中齿轮传动的作用力,建立了一种包含转动惯量、间隙和重合度的齿轮传动接触／冲击作用力模型。然后,利用阀值函数将其化简为可参数估计的模型结构,采用最小二乘迭代估计法对模型参数进行估计,并通过齿轮传动实验验证了模型的合理性。针对锻造操作机夹持系统并联驱动的一些非线性因素,建立了夹持系统并联驱动的全局动力学模型,并通过实验验证了模型的合理性。针对系统动态特性差异的主要因素,如油压差异、阀芯行程差异、马达排量差异以及齿轮的刚度系数和间隙的差异等进行了均载分析研究,分析结果显示系统的动态特性差异对并联驱动均载性造成不同程度的影响。根据建立的夹持系统并联驱动模型,构建基于系统模型的近似内模控制器和扭矩误差补偿反馈控制方法进行并联驱动的均载控制研究,并针对系统动态特性差异主要因素的影响进行均载控制仿真分析和实验研究,实验结果表明该控制方法满足夹持系统并联驱动的均载控制要求。实际中并联驱动系统的复杂性和一些不确定非线性因素会造成系统建模误差,所以采用神经网络模型对系统模型进行补偿,并通过实验修正神经网络近似模型,建立了由神经网络近似内模控制器和扭矩误差补偿反馈的均载控制策略,控制仿真结果表明该控制策略有效保证了均载和快速启停定位的控制要求。更多还原

【Abstract】 Heavy duty forging manipulator is a large production equipment with the free forging press to forge the forging pieces. Under gripping the hundreds of tons of forging piece with four meters diameters, the heavy duty forging manipulator’s single tongs rotation systems can’t drive the large forging pieces, so it needs using the parallel driving systems to realize the rotation of forging piece. However, the kinetic difference of driving parts, backlash and friction between parts for the parallel driving systems of forging manipulator causes the unbalance loading, force distortion and other undetermined behaviour of force, reducing the working life of the parallel driving systems, and causing the driving disabler. So it needs further studying the load balance of the parallel driving systems.Because forging manipulator assistants with free forging press to realize coordinating motion, the parallel driving systems need discontinuous rotary motion under the start-stop frequently and quickly. The gear transmission is in the intermittent driving with rectangular impulse under the gear impacting states, which is more serious because of gear backlash. Moreover, the dynamical variation of the parallel driving systems such as interaction force of gear transmission is difficulty to be measured directly from the experiment. So it needs an effective calculation method to solve the gear contact/impact dynamics equation to analyze the gear impact dynamics behaviour under the intermittent driving. In addition, because there are large inertia and variation of load, strong nonlinear charactreristics of hydraulic systems, dead zone of the control valves, gear backlash, friction nonlinear in the parallel driving systems of forging manipulator, influncing directly the position and load balance of the parallel driving systems of forging manipulator under the time variation heavy conditions, so it needs load balance analysis and load balance control for the parallel driving systems of forging manipulator.Therefore, the studies on the parallel driving systems of forging manipulator are conducted in the present paper as follows:A momentum iterative solution approach based on Newmark direct integration method is proposed for the gear contact/impact problem under the intermittent rotation. This solution method can convert the conventional second-order equation of motion in force equilibrium into another first-order momentum equation and further smooth out the rapid changes of discontinuous rotation produced by start-stop frequently. The validation of accuracy and convergence using momentum analytical approach based on Newmark direct integration method is demonstrated by simulating the contact/impact of intermittent gear transmission using3D finite element dynamical software and experiments.A gear transmission interaction force model is proposed to analyze the heavy duty and intennittent gear transmission interaction force characteristics of forging manipulator based on the contact model of robotic system. The inertia, gear backlash and contactratio are introduced into the proposed model, which is simplified to the identification model using an appropriate switching function. Furthermore, a developed least square based iterative parameter estimation method is introduced to estimate some parameters of the interaction force model. The relevant experiment is implemented to validate the simulation results of the proposed model.For many complexity and nonlinearity factors of the actual tongs parallel rotation systems of forging manipulator, the parallel driving dynamics model of the forging manipulator is made based on the single rotary driving system. The load balance and position of forging manipulator is simulated and analyzed for the dynamical difference factors of forging manipulator such as oil pressure, valve friction and motor displacement of hydraulic systems, stiffness coeffiencient and backlash of gear transmission. The simulation results indicat that these factors cause different influence significantly for the load balance or position of the parallel driving systems of forging manipulators.The internal model control and the torque error compensation feedback control are employed for the load balance control of forging manipulators The dynamical difference factors of forging manipulator are analyzed by the simulation and experiment of load balance control, it indicates that the load balance control strategy meet basicly the load balance control requirement that it realize load balance and start-stop quickly simultaneously for the parallel driving of forging manipulator gripping systems.In fact, many complexity and nonlinearity factors of the actual tongs parallel rotation systems of forging manipulator can produce error of system model. In oder to further improve the load balance control performance with start-stop quickly, the approximate internal model based neural network is updated by the experiments. Then the approximate internal model-based neural control (AIMNC) and the torque error compensation feedback control are employed for the load balance control experiment of forging manipulators. Through the simulation analysis, the control stratagem is validation for the load balance control.更多还原