直线电机系统的复合智能控制

发布时间：2018-12-23 20:44

【摘要】：和传统的旋转电机相比较,直线电机由于没有中间传动装置的特点,使得其具有结构简单、摩擦小、灵敏度高、精度高、速度大小不受到限制等独特的优势。近年来伴随着直线电机制造水平和相关控制技术的发展,直线电机已被广泛地应用于军事武器、工业生产以及生活用品的各个领域,例如磁悬浮列车、高速高精加工、芯片和集成电路制造、高精密仪器等。随着制造业日益朝着高速化、精密化趋势的发展,将来直线电机在生产制造以及日常生活中的应用必然会更加广泛。因此,研究直线电机结构原理和直线电机运动控制技术就显得十分有意义。PID控制技术是一种具有悠久历史的控制技术,凭借操作简单、易实现等优点,现已成为当今工业控制中应用最为普遍也最为成熟的控制方法。众所周知,PID控制器中含有三个需要调节的控制参数(比例、积分、微分系数),实际生产使用中若想得到满意的控制效果就必须对PID控制器的这三个参数进行调节。目前实际生产中PID参数的整定方法大部分为手工整定、基于经验或者基于整定公式算法的整定方法。然而,由于实际的工业控制过程往往是多样化、非线性和瞬变性的,这些方法整定的PID参数系统控制效果往往不能满足要求,尤其对于速度和精度要求都较高的直线电机系统PID控制器整定的效果更是差强人意。本文在建立了直线电机的数学模型基础之上,分析了各种PID控制器参数的整定方法,针对实验永磁直线电机的模型和直线电机系统本身的特点,设计了直线电机的PID控制器,提出利用果蝇优化算法优化直线电机BP神经网络的权值矩阵来实现PID参数的自整定,对实验直线电机建立Matlab/Simulink模型,进行仿真,并将其和传统BP神经网络PID控制参数自整定的控制结果进行比较。传统的直线电机伺服系统通常采用PID位置跟踪反馈控制的方式,但是随着对控制精度的要求不断提高,一般的PID控制由于控制精度的限制、负载扰动等因素的影响,控制效果往往不能达到预期效果。本文针对这种情况提出了自适应“速度+加速度”前馈补偿的复合前馈PID控制策略,即保留了原有PID控制的特性又能充分利用速度前馈和加速度前馈控制的优点,使直线电机能够更好地应用于高速、高精的运动控制中。最后,利用MATLAB/Simulink对实验系统的直线电机建立了带有“速度+加速度”自适应复合前馈的模型,并对直线电机模型的升降速曲线控制效果进行仿真。利用计算机控制的直线电机实验系统进行大量实验,验证了结论的正确性,实现了直线电机系统的复合智能控制。
[Abstract]:Compared with the traditional rotary motor, the linear motor has the advantages of simple structure, small friction, high sensitivity, high precision and no limitation of speed because it has no intermediate transmission. In recent years, with the development of linear motor manufacturing level and related control technology, linear motor has been widely used in military weapons, industrial production and daily necessities, such as maglev train, high-speed and high-precision processing. Chip and integrated circuit manufacturing, high-precision instruments, etc. With the development of manufacturing industry towards high speed and precision, the application of linear motor in production and daily life will be more and more extensive in the future. Therefore, it is of great significance to study the principle of linear motor structure and the motion control technology of linear motor. PID control technology is a control technology with a long history, with the advantages of simple operation and easy realization, etc. Nowadays, it has become the most common and mature control method in industrial control. As we all know, there are three control parameters (proportion, integral, differential coefficient) in PID controller. In order to obtain satisfactory control effect, the three parameters of PID controller must be adjusted. At present, most of the tuning methods of PID parameters in actual production are manual setting, based on experience or based on tuning formula algorithm. However, because the actual industrial control process is often diversified, nonlinear and transient, the control effect of the PID parameter system adjusted by these methods is often unable to meet the requirements. Especially for linear motor system with high speed and precision, the effect of PID controller tuning is not satisfactory. On the basis of establishing the mathematical model of linear motor, this paper analyzes the setting methods of various parameters of PID controller. According to the model of experimental permanent magnet linear motor and the characteristics of linear motor system, the PID controller of linear motor is designed. The weight matrix of BP neural network of linear motor is optimized by Drosophila optimization algorithm to realize the self-tuning of PID parameters. The Matlab/Simulink model of experimental linear motor is established and simulated. It is compared with the traditional BP neural network PID control parameters self-tuning control results. The traditional linear motor servo system usually adopts the PID position tracking feedback control method, but with the continuous improvement of the control precision, the general PID control is affected by the control precision limitation, load disturbance and other factors. Control effects often fail to achieve the desired results. In this paper, a compound feedforward PID control strategy based on adaptive "velocity acceleration" feedforward compensation is proposed, which preserves the characteristics of the original PID control and makes full use of the advantages of the speed feedforward control and the acceleration feedforward control. The linear motor can be better used in high speed and high precision motion control. Finally, the model of linear motor with "velocity acceleration" adaptive compound feedforward is established by using MATLAB/Simulink, and the control effect of the ascending and descending curve of linear motor model is simulated. A large number of experiments are carried out by using the computer controlled linear motor experimental system, which verifies the correctness of the conclusion and realizes the compound intelligent control of the linear motor system.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM359.4;TP273.5

【相似文献】