基于逐层预测模型的感应电机效率优化滑模控制

发布时间：2018-08-14 20:23

【摘要】：针对轻载时感应电机效率优化运行问题,减少转子磁链是一种有效的方法,但转子磁链受电动机参数变化的影响,为了切实保证感应电机运行效率最优,在不同的工况下得到合适的转子磁链给定值成了问题的关键。为了提高转子磁链的预测精度,本文提出了一种基于二叉树型分层神经网络的逐层预测模型。该模型通过逐层细化预测范围并选用多个相同结构的神经网络进行递推,在复杂的工况下得到合适的转子磁链给定值,实现系统效率最优。此外,为了提高效率优化过程中电机转速的响应速度,设计了一种基于新型趋近律的滑模变结构控制策略,从而保证系统轻载时运行效率高并且转速响应迅速。仿真和实验验证了方案的有效性。
[Abstract]:Reducing rotor flux is an effective method to optimize the efficiency of induction motor under light load. However, the rotor flux is affected by the variation of motor parameters, in order to ensure the optimal operation efficiency of induction motor. The key to the problem is to obtain the appropriate rotor flux value under different working conditions. In order to improve the prediction accuracy of rotor flux, a hierarchical prediction model based on binary tree type hierarchical neural network is proposed in this paper. In this model, the prediction range is refined layer by layer and multiple neural networks with the same structure are selected for recursion. The given rotor flux is obtained under complex working conditions, and the system efficiency is optimized. In addition, a sliding mode variable structure control strategy based on a new approach law is designed to improve the speed response of the motor in the process of efficiency optimization, so as to ensure the high efficiency of the system under light load and the rapid response of the speed. The effectiveness of the scheme is verified by simulation and experiments.
【作者单位】：河北工程大学信息与电气工程学院;
【基金】：国家自然科学基金(11272112) 河北省科技攻关(12213912D)资助项目
【分类号】：TM346

【参考文献】