【摘要】 为解决现有机理模型参数多且取值误差大导致的模型精度低、数据模型构建简易不能揭示钻削机理等问题，开展了基于机理与数据驱动的CFRP钻削轴向力建模研究。通过对现有CFRP钻削力模型进行修正，得到适用性更广、精度更高的轴向力机理模型。基于神经网络和实验数据构建轴向力机理模型预测偏差补偿器，采用钻削机理与钻削数据联合驱动的方式建立了CFRP钻削轴向力模型。经验证，新建立的模型可在机理模型的基础上进一步提升预测精度，在样本数据集和随机数据集中，预测平均相对误差分别为机理模型的1/8和1/3。更多还原

【Abstract】 In order to solve these two problems, the existing mechanism model has low accuracy due to multiple parameters and large parameter value error, and the data model is simple to build but cannot reveal the drilling mechanism, a model of CFRP drilling axial force based on mechanism and data driving is carried out in this paper.The existing CFRP cutting force model is modified to obtain the thrust force mechanism model with wider applicability and higher accuracy.Based on neural network and experimental data, a compensator for prediction deviation of thrust force mechanism model is constructed.The thrust force model of CFRP drilling is established by combining drilling mechanism and drilling data.It is proved that the new model can further improve the prediction accuracy based on the mechanism model: the mean relative error of prediction is reduced by 8 times in the sample data set and 3 times in the random data set, respectively.更多还原