基于自抗扰控制技术的永磁同步电机调速系统

发布时间：2018-02-12 08:06

  本文关键词： 自抗扰控制器 转动惯量辨识 负载转矩观测 辨识补偿 转速估计　出处：《西南交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：永磁同步电机具有高功率密度,体积小,散热好等一系列优点。目前,永磁同步电机已在高精度、高速伺服领域扮演重要的角色。永磁同步电机得到广泛应用的同时也对电机控制性能提出更高的要求,尤其对控制系统的抗扰性能提出更大的挑战。为抑制负载变化给电机输出转速带来的影响,本文对基于自抗扰控制技术的永磁同步电机控制策略进行研究,采用了一种Gopinath观测器补偿自抗扰控制器的方案。对转速环和电流环采用自抗扰控制器的永磁同步电机矢量控制系统进行分析,电机控制问题中自抗扰控制技术相对于传统PI控制具有更好的动态响应性能和对内外扰动抑制效果。本文对于自抗扰控制器参数整定问题:对线性结构自抗扰控制器的控制参数整定,通过推导其系统闭环特征方程完成。对非线性结构自抗扰控制器的控制参数整定,采用免疫粒子群算法完成。为减轻扩张状态观测器负担,进一步提高控制系统抗扰性能,本文对传统自抗扰控制器进行辨识补偿,辨识补偿的物理量包括永磁同步电机的转动惯量和负载转矩。转动惯量辨识方面,分析对比了三种在线辨识方法:带遗忘因子的最小二乘法、基于波波夫超稳定性的模型参考自适应、基于郎道离散迭代的模型参考自适应算法。通过选择合理的增益可以准确地在线辨识转动惯量。负载转矩观测方面,常用的在线观测方法采用降阶观测器算法对状态量进行重构,本文根据所采用的方案特点,采用Gopinath观测器对负载转矩进行观测,通过稳定性分析选择合理控制参数可以准确对负载转矩进行观测。对于不适合安装转速编码器的特殊场合,本文采用模型参考自适应算法实现基于自抗扰控制技术的永磁同步电机无速度传感器控制,使用的无速度传感器控制算法结构简单易实现,系统性能可满足基本要求。实验部分采用PESX-Ⅱ电力电子与电力传动实验平台,以TMS320F2812为控制核心。对采用PI控制、自抗扰控制、辨识补偿自抗扰控制的永磁同步电机控制方案进行验证,通过分析对比实验结果与仿真结果来验证采用辨识补偿自抗扰控制方案的有效性和优越性,方案可应用于实际控制中。
[Abstract]:Permanent magnet synchronous motor (PMSM) has a series of advantages such as high power density, small size, good heat dissipation and so on. The field of high speed servo plays an important role. The permanent magnet synchronous motor (PMSM) has been widely used, and the control performance of PMSM is also required to be higher. In order to restrain the influence of load variation on motor output speed, the control strategy of PMSM based on ADRC is studied in this paper. A scheme of Gopinath observer compensation ADRC is adopted. The vector control system of permanent magnet synchronous motor (PMSM) with active disturbance rejection controller in speed loop and current loop is analyzed. The ADRC technique in motor control problem has better dynamic response performance compared with the traditional Pi control, and it can restrain the internal and external disturbance. In this paper, the parameter tuning problem of the ADRC controller and the linear structure ADRC control are discussed. The control parameters of the machine are set, By deducing the closed-loop characteristic equation of the system, the immune particle swarm optimization algorithm is used to adjust the control parameters of the nonlinear structure ADRC. In order to reduce the burden of the extended state observer, the disturbance rejection performance of the control system is further improved. In this paper, the traditional ADRC is identified and compensated. The physical parameters of identification compensation include moment of inertia of permanent magnet synchronous motor, load torque and moment of inertia identification. Three online identification methods are analyzed and compared: least square method with forgetting factor and model reference adaptation based on Popov hyperstability. A model reference adaptive algorithm based on Lang Dao discrete iteration is proposed. The moment of inertia can be accurately identified online by selecting a reasonable gain. The commonly used on-line observation method uses the reduced order observer algorithm to reconstruct the state quantity. According to the characteristic of the scheme, this paper uses the Gopinath observer to observe the load torque. The load torque can be accurately observed by selecting reasonable control parameters through stability analysis. In this paper, the speed sensorless control of PMSM based on ADRC is realized by using model reference adaptive algorithm. The speed sensorless control algorithm used in PMSM is simple and easy to realize. The performance of the system can meet the basic requirements. In the experiment part, PESX- 鈪，

本文编号：1505162