Abstract:

A modeling method of fuzzy neural network based on neural fuzzy control theory was applied in thermal error  modeling of NC machine tool and the structure and modeling principle of fuzzy neural network on thermal errors was discussed. The experiment on spindle head of large NC gantry guide grinder was conducted and two independent sets of  experimental data were obtained by measuring the temperature of spindle head and the thermal error of spindle with non-contact infrared thermometer and dial gauge, one was used to establish thermal error fuzzy neural network prediction model of  spindle head  and the other was used to validate the model. The test results show that fuzzy neural network model has high prediction accuracy and good generalization. Compared with radial basis gunction(RBF) neural network modeling method, fuzzy neural network modeling method has better modeling efficiency, robustness and predict performance.

Key words: Takagi-Sugeno type fuzzy inference, membership function, fuzzy neural network, robustness, generalization ability

摘要：

将基于神经模糊控制理论的建模方法——模糊神经网络建模法应用到数控机床热误差建模当中，讨论了热误差模糊神经网络的结构及建模原理；对大型数控龙门导轨磨床主轴箱系统进行建模试验，采用非接触式红外温度测量仪和千分表分别测量主轴箱系统温度值与主轴热误差，得到两组独立的试验数据,一组用来建立主轴箱系统热误差模糊神经网络预报模型，另一组用来对模型进行验证。试验结果表明，模糊神经网络模型预测精度高,泛化能力强;将模糊神经网络建模方法与径向基函数神经网络建模方法进行综合对比,分析结果表明,模糊神经网络建模方法具有更好的建模效率、建模鲁棒性及预测性能。

关键词: Takagi-Sugeno型模糊推理, 隶属度函数, 模糊神经网络, 鲁棒性, 泛化能力