基于偏最小二乘回归(PLSR)方法的铣削力模型系数辨识研究

在已有的铣削力模型基础上,建立一种适应多种工况条件下的铣削力系数多项式模型,提出一种方便、有效的多项式模型系数的辨识方法。即通过混合位级正交设计方法设计实验,获得实验数据;利用偏最小二乘回归分析方法——PLSR在模型参数辨识中的优势。仿真和实验结果表明,该方法不但可以准确、可靠地建立适用于不同工况条件的铣削力系数模型,同时,具有实验次数少,建模计算量小、辨识速度快,模型预测能力强等特点,模型参数具有较为明显的物理意义。文中最后给出了铣削加工LY12和45#钢的两种不同材料的仿真实例。

Study on the Identification of the Milling Force Parameter Model Based on Partial Least Square Regression and Application

Based on the typical milling force model, the polynomial milling force parameter model is built in the different cutting conditions. Meanwhile, the fast and effective identification method of model coefficients is provided, which combines the virtues of both orthogonal design theory and partial least-square regression (PLSR) method. The results of the simulations and experiments prove that this kind of method not only build the model of milling force coefficient correctly and effectively but also have the advantages of fewer experiments, less amount of calculation, stronger ability of the model forecast. The coefficients of model have more actual meaning. The two successful applications of the materials of LY12 and 45# steel are given at the end of the paper.