基于神经网络的无轴承电机悬浮子系统仿真研究

Study of the Levitation Subsystem of Bearingless Motors Based on Neural-Network

无轴承电机电磁转矩和磁悬浮力的独立控制是实现电机超高速运转的有效途径。论文在研究电机磁悬浮机理的基础上,推导了磁悬浮力的数学模型,提出了1种基于DRNN神经网络整定的PID控制方法,将该控制方法应用于转子悬浮控制子系统中,并应用MATLAB工具进行了系统仿真。仿真结果表明,该控制系统具有很好的性能,能够实现电机的稳定悬浮。

The separate control of levitation force troque in bearingless motors is very improtant for the high speed of the motors. This paper introduces the operation principle of bearingless motors, and derives the math model of levitation force, then a control model that based on DRNN is put forward and simulated by using MATLAB. The simulation results show that the control system has good performance and can achieve the stable operation of the rotor in bearingless motors.