基于混合遗传算法的摩擦参数辨识研究

发布时间：2019-03-03 08:51

【摘要】：随着各类机电系统对精度要求的不断提升,摩擦环节的存在已经成为提升系统性能的瓶颈。消除摩擦负面影响的有效手段,是在建立摩擦数学模型的基础上完成参数的有效辨识并实现后续摩擦补偿。但是由于摩擦的产生机理较为复杂,而且摩擦过程中往往伴随着受迫振动和间隙碰撞等额外因素的影响,这进一步加大了摩擦参数辨识的难度。本文首先对摩擦模型以及辨识方法的发展历程和研究现状做了简要回顾与分析。随后,在对摩擦机理和摩擦特性进行分析研究的基础上,将滑块沿滑轨自由运动时的摩擦过程作为研究对象,自主研制了考虑运动体自身振动和间隙碰撞的摩擦发生装置。基于摩擦过程和测试数据特性,将正弦滤波项引入经典摩擦模型,建立了修正摩擦模型。最后考虑到摩擦模型是速度的函数,且参数辨识对速度信息的精度有较高的要求,本文采用基于高速摄像机的非接触式测试方法获取滑块的速度信息,并进行后续的数据处理。实验结果验证了模型修正及辨识方法的有效性。本文通过对辨识算法进行特性分析,将模拟退火思想引入遗传算法,并将混合算法应用于实例分析及摩擦参数辨识研究。辨识结果表明,采用本文的修正模型和混合算法,可以获取更为精确的辨识结果。该方法为摩擦特性的进一步研究提供了重要的参考依据。
[Abstract]:With the continuous improvement of precision requirements of various electromechanical systems, the existence of friction links has become the bottleneck to improve the performance of the system. The effective method to eliminate the negative effects of friction is to effectively identify the parameters and realize the subsequent friction compensation on the basis of establishing the mathematical model of friction. However, the mechanism of friction is more complex, and the friction process is often accompanied by forced vibration and clearance impact, which further increases the difficulty of identification of friction parameters. In this paper, the development history and research status of friction model and identification method are briefly reviewed and analyzed. Then, based on the analysis of the friction mechanism and the friction characteristics, the friction process of the slider moving freely along the slide rail is taken as the research object, and the friction generating device considering the vibration and gap collision of the moving body is developed. Based on the friction process and the characteristics of the test data, the sine filter term is introduced into the classical friction model, and the modified friction model is established. Finally, considering that the friction model is a function of velocity and the precision of velocity information is required by parameter identification, the speed information of slider is obtained by non-contact testing method based on high-speed camera, and the subsequent data processing is carried out. The experimental results show that the model modification and identification method is effective. In this paper, the characteristics of the identification algorithm are analyzed, and the idea of simulated annealing is introduced into the genetic algorithm, and the hybrid algorithm is applied to the case analysis and the identification of friction parameters. The identification results show that more accurate identification results can be obtained by using the modified model and the hybrid algorithm in this paper. This method provides an important reference for the further study of friction characteristics.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP18

【相似文献】