基于遗传算法的直流伺服系统参数辨识及摩擦补偿控制

针对静态摩擦力对数控机床直流伺服系统的干扰问题,提出了一种先利用遗传算法对静态摩擦模型中的参数进行辨识,然后采用基于摩擦模型补偿的伺服控制方法。该方法首先根据直流伺服系统的摩擦特性建立摩擦模型,再将摩擦补偿引入到直流伺服系统的反馈控制结构中,获取伺服电机的位置误差。采用遗传算法对摩擦补偿模型进行参数辨识,使摩擦补偿量在数值上不断逼近实际的摩擦干扰,并利用摩擦补偿量来抵消摩擦给伺服系统带来的影响。为了验证参数辨识的效果,将普通PD控制与基于摩擦补偿的PD控制进行了仿真比较,实验结果表明,后者能够消除由于静摩擦的存在而造成的位置跟踪中出现的平顶现象,能够达到理想的跟踪效果。因此,本文所提出的方法具有较强的摩擦干扰补偿能力,能够实现对直流伺服系统的精确控制。

Parameter Identification of the DC Servo Systems and Friction Compensation Control Based on Genetic Algorithms

According to the problem of static friction disturbances in the direct current(DC) servo system of CNC,a servo control method based on friction model compensation after the parameter identification of the static friction model using Genetic Algorithm(GA) is proposed.Firstly,according to the friction characteristics of the DC servo systems,a friction model is built,then the friction compensation into the feedback control structure of the servo systems to obtain the position error of the servo motor is introduced.GA is adopted to identify the parameters of the friction compensation model,and to make the quantity of the friction compensation approximates to the practical friction disturbance unceasingly,so that the influence of friction on the servo systems is eliminated.In order to verify the effect of parameter identification,a comparison is made between the traditional PD control and the PD control based on friction compensation.The simulation results show that the latter can eliminate the flat phenomenon in position tracing caused by the static friction and achieve the desired tracking effect.Therefore,this method has strong ability to compensate the friction interference and can achieve precise control of the DC servo systems.