【作者】 王建明；

【导师】 刘震；

【作者基本信息】 东北大学 ， 电气工程， 2023， 硕士

【摘要】 国民经济重点产业的转型升级、战略性新兴产业的培育壮大和能源资源环境的约束,对智能制造装备行业提出了更高的要求。然而,智能制造产业升级离不开伺服系统的发展,在工业现场,用户需求的智能化对产品的生产质量和效率提出了更高的要求,因此伺服系统的响应性能和控制精度也将面临更加严峻的挑战。以工业自动化程度较高的工业机器人为例,伺服系统作为工业机器人的核心部件,其性能往往也决定着整个系统性能的好坏。工业机器人由于在运动过程中会因其姿态变化进而导致转动惯量变化,会引发伺服系统控制器参数与转动惯量不匹配的问题,甚至会破坏系统的稳定性。为解决伺服系统控制器参数与转动惯量不匹配的问题,本文提出了基于转动惯量辨识的控制器参数自整定方法。首先是对伺服系统运行过程中的转动惯量进行辨识,依据辨识到的结果对伺服系统控制器参数进行整定,进而实现伺服系统性能提升的目标。自整定技术可以使得伺服系统运行过程中,快速获取所需要的控制参数,当伺服系统的被控对象发生变换时,也可以在线对控制器参数进行修正,进而保证伺服系统的稳定性,因此,自整定是完成伺服系统控制器参数优化的有效方法。为了实现对转动惯量进行快速且精准辨识的目标,本文基于扩展卡尔曼滤波算法提出了一种改进的方法。首先是基于扩展卡尔曼滤波算法对转动惯量进行辨识,设计了在不同工况下的仿真实验,依据实验结果总结了扩展卡尔曼滤波算法的改进方向。然后,针对传统的扩展卡尔曼滤波算法存在的系统噪声矩阵Q与测量噪声矩阵R与测量系统不匹配,收敛速度较慢,辨识精度较差等问题,在扩展卡尔曼滤波算法中的预测误差协方差矩阵中,引入了一个修正因子。通过该修正因子不仅可以平衡状态方程的预测值和测量值的权值,对于系统模型对外界的扰动,也会起到调节作用,进而保证算法处于收敛状态。然后进行仿真对比实验,实验结果表明,经过改进后的扩展卡尔曼滤波算法相比于改进前,在进行转动惯量辨识时,无论是收敛速度还是辨识精度,都有了明显提升。伺服系统在工作时,往往需要跟踪不同的位置轨迹。为了验证自整定方法的有效性,将辨识到的转动惯量用于伺服系统控制器参数的自整定,设计了在不同形式的给定信号下,速度与位置控制器在参数自整定前后的速度与位置响应的仿真实验。实验结果表明,控制器参数经过自整定后,实现了与转动惯量相匹配的目标,进一步地,提升了伺服系统的性能。更多还原

【Abstract】 The transformation and upgrading of key industries of national economy,the cultivation and expansion of strategic emerging industries and the constraints of energy,resources and environment have put forward higher requirements for the intelligent manufacturing equipment industry.The improvement and upgrading of intelligent manufacturing industry cannot be separated from the development of servo system.In the industrial field,the intelligent demand of users has put forward higher requirements for the production quality and efficiency of products,so the response performance and control accuracy of servo system will also face more severe challenges.Taking industrial robots with high degree of industrial automation as an example,the performance of servo system,as the core component of industrial robots,often determines the performance of the whole system.The change of the moment of inertia will result from the change of the attitude of the industrial robot in the process of movement,which will cause the mismatch between the controller parameters of the servo system and the moment of inertia,and even destroy the stability of the whole system.In order to solve the problem of mismatch between controller parameters and moment of inertia of servo system,a self-tuning method of controller parameters based on moment of inertia identification is proposed in this paper.Firstly,the moment of inertia in the operation process of the servo system is identified,and the controller parameters of the servo system are set according to the identified results,so as to achieve the goal of improving the performance of the servo system.When the controlled object of the servo system is transformed,the controller parameters can also be modified online to ensure the stability of the servo system.Therefore,the self-tuning is an effective method to complete the optimization of the controller parameters of the servo sy stem.In order to realize rapid and accurate identification of moment of inertia,an improved method based on extended Kalman filter algorithm is proposed.Firstly,the moment of inertia is identified based on the extended Kalman filter algorithm,and the simulation experiments under different working conditions are designed.Based on the experimental results,the improvement direction of the extended Kalman filter algorithm is summarized.Then,in view of the problems existing in the traditional extended Kalman filter algorithm,such as the mismatch between the system noise matrix Q and the measurement noise matrix R and the observation system,slow convergence rate and poor identification accuracy,a correction factor is introduced into the prediction covariance matrix of the extended Kalman filter algorithm.Through this correction factor,the predicted value and the observed value weight of the equation of state can be balanced.It also plays a regulating role in the disturbance of the system model to the outside world,so as to ensure that the algorithm is in a convergence state.Then the simulation and comparison experiments are carried out.The experimental results show that the improved extended Kalman filter algorithm has significantly improved both the convergence speed and the identification accuracy in the moment of inertia identification.When the servo system is working,it often needs to track different position trajectory.In order to verify the effectiveness of the self-tuning method,the identified moment of inertia is used to self-tune the controller parameters of the servo system,and the simulation experiment of the response of the speed and position controller before and after the parameter self-tuning is designed under different forms of given signals.The experimental results show that the controller parameters can match the moment of inertia after self-tuning,and further improve the performance of the servo system更多还原