电主轴定子电阻混合智能辨识方法

目的研究不同工况下电主轴定子电阻变化规律，找寻有效的定予电阻辨识方法，提高电主轴直接转矩控制（DTC）系统性能和定子电阻辨识精度．方法实验检测电主轴不同工况下定子电阻，分析影响定子电阻的主要因素，将影响定子电阻变化的变量作为辨识系统的输入，建立人工神经元网络（ANN）与案例推理（CBR）联合组成的混合式智能定子电阻辨识的模型．结果单纯地通过人工神经元网络方法对定子电阻进行辨识，辨识的误差可以达到±0．005Q；利用案例推理技术对神经元网络的辨识结果进行修正后，定子电阻辨识的误差减小到±0．002n．结论基于神经元网络及案例推理相结合的定子电阻辨识方法与单纯的神经元网络辨识相比，可将定子电阻辨识精度提高60％，将辨识以后的定子电阻导入到电主轴直接转矩控制系统中，大大地提高了系统的控制性能．

The Method about Stator Resistance Hybrid Intelligent Identification of Motorized Spindle

In order to improve the precision of the stator resistance identification, and solve the influence of stator resistance change on the performance of the motorized spindle in direct torque control （DTC）, based on artificial neural network（ANN） and Case-Based Reasoning （CBR）, a hybrid intelligent stator resistance i- dentification method is introduced. The experiments are used to obtain the data needed by stator resistance identification and determined the relationship between main factors and the stator resistance. The simulation is used to prove the approach. The results confirm that the identification error is ± 0. 005 Ω by using neural network, but the identification error is _＋ 0. 002 Ω by using ANN-CBR. Studies show hybrid intelligent meth- od based on ANN-CBR can obtain higher recognition accuracy for the stator resistance. It is helpful to im- prove the performance of direct torque control.