基于改进MRAS异步电机矢量控制系统转速辨识的研究

发布时间：2018-06-03 06:44

本文选题：异步电机矢量控制 + 转速辨识　；参考：《哈尔滨工业大学》2015年硕士论文

【摘要】：异步电机作为一种结构简单,制造工艺难度较低,节省成本及运行更加安全可靠的动力机械,在各个行业都有着广泛的应用。然而,随着研究的不断深入和实际情况的需要,基于矢量控制原理,对于异步电机构成的交流传动系统或交流伺服控制系统要求越来越高。利用转速辨识结构来替代位置传感器,既能提升系统可靠性,也能降低硬件成本。然而,取代位置传感器的前提是需要准确的转速辨识,本文围绕电机的转速辨识进行以下几个方面的研究:从电机的数学模型出发,根据矢量控制原理,用Simulink搭建了基于转子磁场定向的异步电机矢量控制系统仿真平台,并在此平台上进行了仿真。阐述了参数辨识的方法,研究模型参考自适应(MRAS)算法,推导出基于波波夫超稳定性理论的转速辨识自适应律。在MRAS算法的基础上,提出了一种基于消积分饱和的改进型算法,有效地消除了电压纯积分环节引起的直流偏差、积分饱和等影响。并对改进算法构成的整个无位置传感器控制系统进行了小信号分析与建模,依据根轨迹判定出系统的稳定性,仿真结果表明基于改进MRAS法构成的闭环控制系统在给定的转速以及全负载范围内是稳定的。此外针对改进MRAS速度辨识系统进行了参数敏感性分析,仿真表明电机参数变化会对辨识收敛速度产生影响。对MRAS及其改进型算法进行了转速、负载和转矩突变等试验的对比仿真和分析,仿真结果表明改进型算法的辨识准确度更高和超调量更小。搭建出整个实验平台,设计并焊接电压电流采样电路。在以XMC4500处理器为核心,以DAVE 3为软件开发平台,设计并实现整个算法程序,最终在搭建的实验平台上验证了MRAS及其改进型算法,实验结果表明改进型算法的转速辨识具有超调更小,响应时间更快以及准确度更高等特点。
[Abstract]:As a kind of power machinery with simple structure, low difficulty in manufacturing process, cost saving and more safe and reliable operation, asynchronous motor has been widely used in various industries. However, with the deepening of the research and the need of the actual situation, based on the principle of vector control, the requirements of AC drive system or AC servo control system composed of asynchronous motor are becoming more and more high. The speed identification structure is used to replace the position sensor, which can not only improve the system reliability, but also reduce the hardware cost. However, the premise of replacing the position sensor is that accurate speed identification is required. This paper focuses on the speed identification of the motor in the following aspects: from the mathematical model of the motor, according to the principle of vector control, The simulation platform of induction motor vector control system based on rotor flux orientation is built with Simulink, and the simulation is carried out on this platform. This paper describes the method of parameter identification, studies the model reference adaptive algorithm, and deduces the speed identification adaptive law based on Popov's hyperstability theory. On the basis of MRAS algorithm, an improved algorithm based on elimination of integral saturation is proposed, which effectively eliminates the effects of DC deviation and integral saturation caused by voltage pure integral. The whole position sensorless control system composed by the improved algorithm is analyzed and modeled with small signal, and the stability of the system is determined according to the root locus. The simulation results show that the closed-loop control system based on the improved MRAS method is stable in the given speed and full load range. In addition, the parameter sensitivity analysis of the improved MRAS speed identification system is carried out. The simulation results show that the change of motor parameters will affect the speed of identification convergence. The MRAS and its improved algorithm are simulated and analyzed in the experiments of rotational speed, load and torque mutation. The simulation results show that the improved algorithm has higher identification accuracy and less overshoot. Build the whole experimental platform, design and welding voltage and current sampling circuit. With XMC4500 processor as the core and DAVE 3 as the software development platform, the whole algorithm program is designed and implemented. Finally, the MRAS and its improved algorithm are verified on the experimental platform. The experimental results show that the speed identification of the improved algorithm has the advantages of smaller overshoot, faster response time and higher accuracy.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM343

【相似文献】