【作者】 王强；

【导师】 王淑旺；

【作者基本信息】 合肥工业大学 ， 机械设计及理论， 2024， 硕士

【摘要】 永磁同步电机(Permanent Magnet Synchronous Motor,PMSM)以其卓越的驱动效率、出色的轻量化设计和优异的功率密度特性,在新能源汽车驱动系统中得到了广泛的应用。为确保电机的稳定运行,精准掌握转子的位置信息至关重要。目前,多数系统依赖传感器来获取转子的位置数据。机械传感器不仅需要占据额外的空间,还增加电驱动系统成本,故利用电机自身特性获取转子位置成为重要的研究方向之一。另外,汽车在行驶过程中工况变化频繁,电机受到的负载会频繁发生变化,从而会对电驱动控制系统造成影响,因此含有转矩负载补偿的无位置传感器控制策略有良好的应用前景。首先针对传统的PMSM零/低速段实现无位置传感器控制精度低、转子极性辨识困难的问题,本文改进了传统的高频信号注入法,优化了载波信号的分离和位置误差提取方法代替传统的滤波环节,提出了高频方波信号组合一段正反脉冲的方法实现了高精度的转子极性辨识和位置辨识,并在MATLAB/Simulink平台搭建了仿真模型并进行了分析验证。之后针对转矩波动对传统PMSM控制系统的影响大,控制精度低的问题,本文设计了基于前馈控制的负载转矩观测器对负载转矩进行前馈补偿,分别利用龙贝格观测器和卡尔曼观测器对负载状态进行观测,通过前馈补偿方式实现对负载扰动的抑制。经过仿真验证,上述改进能够有效抑制负载扰动造成波动的现象,提高电机的动态控制性能。最后,本文搭建了以DSP TMS570LS1115为主控芯片的实验平台,在台架上对提出的改进算法进行了试验验证。实验结果表明,本文设计的零/低速无传感器控制精度高,磁极辨识准确度高,负载扰动观测器能有效补偿负载扰动,从而显著改善了电机的控制性能。更多还原

【Abstract】 The Permanent Magnet Synchronous Motor(PMSM)has gained widespread application in the drive systems of new energy vehicles due to its excellent driving efficiency,outstanding lightweight design,and superior power density characteristics.To ensure the stable operation of the motor,accurately grasping the rotor position information is crucial.Currently,most systems rely on sensors to acquire rotor position data.However,mechanical sensors not only occupy additional space but also increase the cost of the electric drive system.Therefore,utilizing the inherent characteristics of the motor to obtain rotor position has become one of the important research directions.Furthermore,as vehicles frequently encounter changing working conditions during operation,the load on the motor varies constantly,potentially affecting the electric drive control system.Hence,sensorless control strategies with motor load compensation have promising application prospects.Firstly,addressing the issues of low precision in traditional sensorless control and difficulty in rotor polarity identification during the zero/low-speed range of PMSM,this paper improves the traditional high-frequency signal injection method.It optimizes the separation of carrier signals and position error extraction methods,replacing traditional filtering techniques.A method combining high-frequency square wave signals with a sequence of positive and negative pulses is proposed to achieve high-precision rotor polarity and position identification.A simulation model is built and analyzed on the MATLAB/Simulink platform.Subsequently,to address the significant impact of torque fluctuations on traditional PMSM control systems and the issue of low control precision,this paper designs a load torque observer based on feedforward control to perform feedforward compensation for load torque.The load state is observed using the Luenberger observer and Kalman observer,and load disturbances are suppressed through feedforward compensation.Simulation verification demonstrates that the improvements can effectively suppress load disturbance-induced fluctuations,enhancing the motor’s dynamic control performance.Finally,this paper establishes an experimental platform using DSP TMS570LS1115 as the main control chip and tests the proposed improved algorithms on a test bench.The experimental results show that the zero/low-speed sensorless control designed in this paper achieves high precision and accurate magnetic pole identification.The load disturbance observer can effectively compensate for load disturbances,significantly improving the motor’s control performance.更多还原