高性能智能化感应电机变频调速系统研究

【作者】 魏彦；

【导师】 黄守道；

【作者基本信息】 湖南大学 ， 电机与电器， 2003， 硕士

【摘要】 本文重点研究了高性能智能化感应电机变频调速系统这一涉及学科范围广泛的课题中一个极为重要的方面——矢量控制系统的全数字实现。本文在分析感应电机数学模型的基础上，详细讨论了转子磁场定向矢量控制理论和空间矢量脉宽调制技术原理，提出了SVPWM的软件实现方法。在系统实现上，重点描述了基于DSP的全数字感应电机矢量控制系统的软硬件实现。在硬件实现中，重点讨论了电流、电压、速度检测单元以及PWM发生单元和系统保护功能等。在软件方面，重点论述了适合于定点DSP的系统变量数字表述形式、电流模型模块和软件空间矢量模块的设计以及PC机控制对象参数预置和修改的在线实现。在分析可能影响全数字调速系统低速性能的软硬件因素之后，给出了改善系统低速性能的对策。同时通过以此控制系统为控制器的系统试验，我们论证了使用高速数字信号处理芯片，在提高控制系统实时性，提高控制系统运行性能以及可靠性和稳定性等方面的可行性。最后，在常规带速度传感器矢量控制系统实现的基础上，本文讨论了无速度传感器矢量控制的仿真实现，提出了一种神经网络与模型参考自适应相结合的转子转速辨识方法。仿真研究表明，应用该神经网络速度估计器的无速度传感器矢量控制系统获得了较为理想的效果。更多还原

【Abstract】 This paper focuses its emphasis on the digital control of induction motor, one of the most important aspects of the high-performance variable-frequency variable-speed system of induction motor. Based on the mathematical model of induction motor, this paper discusses the rotor-field-orientated control principium and the space vector pulse width modulation principle. Subsequently the software design of SVPWM is put forward. The software and hardware design of the DSP-based digital induction motor vector control system are always expounded in the chapter of system realization. The hardware design includes these main circuits such as current sensing, DC bus voltage sensing, space-vector PWM waveform generation, the system protection and so on. In the design of system software, we take account of the numerical format considering a fixed-point DSP TMS320F240, the software designs of the current model and software SVPWM model, and the online realization of parameters setting and changing in PC. After analyzing the software and hardware reasons which can affect the system performance when rotor speed is low, accordingly the improving methods are advanced. According to the system experimentation, it can be concluded that the reliability and stability of the system are all improved by using high-speed digital signals processor. Furthermore, the dynamic performance of the controller is more satisfying. Considering that only the conventional sensored vector control is realized in our system owing to the limitation of the hardware, this paper also discusses the speed-sensorless vector control in virtue of MATLAB/SIMULINK. In that chapter, a estimating method of rotor speed combining the neural network and the model reference adaptive system is brought out. The simulation result indicates that the speed-sensorless vector control system of induction motor using designed neural network estimator can get better performance.更多还原