论文部分内容阅读
针对残余应力和铣削载荷引起的钛合金宽弦空心风扇叶片铣削加工变形问题,提出了一种空心风扇叶片加工变形预测方法。通过正交试验得出铣削力经验公式,采用两种不同的走刀方式,结合“生死单元”技术实现叶片进排气边加工仿真;运用BP神经网络算法建立钛合金宽弦空心风扇叶片加工变形预报模型。通过分析不同走刀方式的加工变形量得出:环切铣削方式对于加工变形的影响大于行切铣削方式。变形预报结果与实际测量值相对误差在10%以内,变形预报相比有限元计算周期短,预测效率高。
In order to solve the problem of milling deformation of titanium alloy wide-chord hollow fan blade caused by residual stress and milling load, a method of deformation prediction of hollow fan blade is proposed. The empirical formula of milling force is obtained through orthogonal experiment. Two kinds of different cutting ways are adopted, and the simulation of blade inlet and exhaust is carried out with the combination of “life and death unit” technology. The BP neural network algorithm is used to build the titanium alloy wide and chord hollow fan blade Machining deformation forecasting model. Through analyzing the processing deformation of different ways, it is concluded that the effect of ring milling method on the processing deformation is greater than the milling method. The relative error between the deformation forecast results and the actual measured values is less than 10%, the deformation forecast is shorter than the finite element calculation period, and the forecasting efficiency is high.