直接转矩控制的异步电动机转矩脉动最小化

Torque ripple minimization in direct torque control of induction motors

针对异步电动机直接转矩控制低速转矩脉动大的问题,充分利用模糊控制吸收人的经验思维,以及神经网络对信息的处理具有自组织、自学习的特点,提出一种新的模糊神经网络控制方法.该方法实现了逆变器开关周期的占空比控制,使感应电动机的转矩脉动达到最小.其中,模糊神经网络的训练采用最小二乘法,解决了常规的BP算法容易陷入局部极小的问题.将传统的直接转矩控制方案和模糊神经网络占空比控制方案进行了比较研究,仿真结果校验了模糊神经网络占空比控制方案的有效性.

As kind of new fuzzy to large torque ripple in direct torque control （DTC） of induction motors at low speed, a neural networks （FNN） approach was proposed based on the merits that fuzzy control absorbs man＇s empirical thinking and neural networks have self-organization and self-study ability. The new approach achieved inverter switch＇s duty ratio control and made torque ripple minimum. The problem that BP algorithm easily gets into local minimum was solved by using least square method for the training of fuzzy neural networks. By comparing traditional DTC approach with FNN duty ratio control scheme, the effectiveness of FNN duty ratio control scheme was verified.