论文总字数：26292字

摘 要

感应电机的多变量控制理论经过多代人的发展，己经日趋成熟并且成功使用于变速矢量控制领域。目前感应电机变速矢量系统的控制方法分类较多，其理论基础也各不相同。在这之中，矢量控制可以说是使用得较为广泛的一类方案。较长时间以来为了对电机控制系统闭环控制，往往会用速度传感器等装置测量实际转速。而速度传感器的安装和其维护是一个较大的问题。同时，也容易受环境的干扰，存在不确定性因素。而无速度传感器理论的出现则能够解决速度传感器由安装及发生故障时系统会受到较大的干扰等等。同时，由于省去了速度传感器这一环节从而获得经济上的优势。

无速度传感器矢量控制理论的目标是如何让转速辨识达到更高的准确程度，这同时也是磁场定向的基础。模型参考自适应法(MRAS)可以说是目前无速度传感器转速辨识理论中比较完善的。MRAS需要有准确的感应电机参数用于参考模型以及可调模型的建立才能使转速辨识达到较高的精确程度。但是电机的实际运行中，温度的升或降会使电机部分重要参数诸如定子电阻以及转子电阻大小也产生改变。从而影响转速估算精度程度，加入部分电机参数在线辨识则可保证转速估算结果达到较高的精度。

基于MRAS理论中的反电势模型的转速辨识能够去除基于磁链模型具有的纯积分环节产生误差累计和直流漂移的影响。但感应电机零速度或极低速度情况下定子电阻将会随温度的上升或下降导致实际数值与给定数值产生差异，从而导致所建立的MRAS模型在感应电机零速度或极低速度运行时转速估算达不到足够的精确度。所以文章采用基于MRAS反电势模型的定子电阻和转速交互式辨识，估算转速以及定子阻值，以此在线调节定子电阻数值实时更新转速以达到更高精确度。

本文依据控制理论建立了全阶磁链观测器以及定子电阻和转速交互式辨识的数学模型。通过Matlab/Simulink的仿真平台，建立了磁链观测器的模型，代入到矢量控制系统中仿真，仿真达到了较高程度模拟现实的要求。其结果说明MRAS能精确估算速度，高精度辨识定子电阻，感应电机矢量控制性能达到本次设计目标。

关键词：感应电机，全阶状态观测器，转速辨识，定子电阻辨识，MATLAB

Analysis and Design of High Performance Flux Observer for Induction Motor

Abstract

The multivariable control theory of induction motors has been mature and successfully used in the field of variable speed vector control after generations of development. At present, the control method of variable speed vector system of induction motor is more and more, and its theoretical basis is different. In this, the vector control can be said to be more widely used in a class of programs. For asynchronous motor vector control, for a long time in order to the motor control system closed-loop control, often with speed sensors and other devices to measure the actual speed. The speed sensor installation and maintenance is a big problem. At the same time, it is also susceptible to environmental interference, there are uncertain factors. And the advent of the speed sensor theory can solve the speed sensor by the installation and failure of the system will be subject to greater interference and so on. At the same time, because of the need to save the speed sensor this link to obtain economic advantages.

The goal of the speed sensorless vector control theory is how to achieve a higher degree of accuracy in speed identification, which is also the basis for magnetic field orientation. Model reference adaptive method (MRAS) can be said that the current speed sensorless speed identification theory is relatively perfect. MRAS requires accurate induction motor parameters for the reference model and the establishment of an adjustable model to achieve a higher degree of accuracy in speed identification. However, the actual operation of the motor, the temperature rise or fall will make some important parameters of the motor, such as stator resistance and rotor resistance size also changes. Its numerical changes will affect the accuracy of the speed estimation accuracy, the need to add part of the motor parameters on-line identification to ensure that the speed estimation results to achieve high accuracy.

Based on the MRAS theory, the rotational speed identification of the back EMF model can remove the influence of the cumulative and DC drift based on the pure integral part of the flux linkage model. However, the stator resistance will cause the actual value to differ from the given value with the temperature rise or fall in the case of zero or very low speed of the induction motor, which leads to the establishment of the MRAS model at the speed of the induction motor zero speed or very low speed Estimates do not reach enough accuracy. So the article based on the MRAS back EMF model of the stator resistance and speed of interactive identification, identification speed and stator resistance, in order to adjust the stator resistance value on-line real-time update speed to achieve higher accuracy.

Based on the control theory, a mathematical model of full-order flux observer and interactive identification of stator resistance and speed is established. Through the simulation platform of Matlab / Simulink, the model of flux observer is established, and the simulation is carried out in the vector control system, and the simulation has reached the requirement of high degree of simulation reality. The results show that MRAS can accurately estimate the speed, high-precision identification of stator resistance, induction motor vector control performance to achieve this design goal.

Keywords： Induction motor, full - order state observer, speed identification, stator resistance identification，MATLAB

目录

摘 要 I

Abstract II

第一章 绪论 1

1.1 研究背景、目的及意义 1

1.1.1 研究背景 1

1.1.2 研究目的 1

1.2 国内外研究现状 2

1.2.1 国外研究现状 2

1.2.2 国内研究现状 3

1.3 论文的安排 4

第二章 感应电机矢量控制理论 6

2.1 感应电机的数学模型 6

2.1.1 三相静止坐标系中的感应电机数学模型 6

2.1.2 感应电机的坐标变换 8

2.1.3 感应电机处于两相静止坐标系时数学模型 9

2.1.4 感应电机处于两相同步速旋转dq坐标系时的数学模型 9

2.2 转子磁场定向 10

2.2.1 直接磁场定向 10

2.2.2 间接磁场定向 11

第三章 转子磁链观测器及交互式转速、定子电阻辨识设计 13

3.1 磁链观测器模型 13

3.1.1状态观测器的基本思想 13

3.1.2 转子磁链观测器模型 14

3.2模型参考自适应法的转速辨识 16

3.2.1模型参考自适应法 16

3.2.2 基于反电势模型的转速辨识 18

3.3 定子电阻辨识 20

3.3.1基于MRAS电机定子电阻与转速交互式辨识 20

第四章 系统仿真和试验结果的分析 22

4.1 仿真工具及仿真参数 22

4.1.1 仿真工具 22

4.1.2 仿真参数 22

4.2 仿真模型 23

4.2.1 坐标变换的仿真模型 23

4.2.2 PI调节器的仿真模型 24

4.2.3 磁链观测器的仿真模型 24

4.2.4 MRAS定子电阻及转速辨识的仿真模型 25

请支付后下载全文，论文总字数：26292字