采用多元非线性回归模型的无头铆钉安装干涉量预测

Predicting Interference of Headless Rivet Using Multiple Nonlinear Regression

为明晰被连接件材料性能对安装干涉量的影响规律,并进一步为新材料扩展应用提供可靠性预测,在铆接过程有限元仿真数据基础上,提出了一种采用多元非线性回归模型的无头铆钉安装干涉量预测方法。首先,根据实际铆接过程建立有限元仿真模型,通过铆接试验验证模型有效性。然后,采用有限元和正交试验法,研究了被连接件弹性模量、屈服强度、强化系数和应变强度指数及其交互作用对铆接干涉量水平的显著性,明确了各因素对安装干涉量的影响效果。最后,选用幂函数作为多元非线性回归的函数形式,剔除显著性较低因素项,建立了待测位置干涉量回归预测模型。结果表明,在一定铆接工艺条件下,屈服强度和应变强度指数及其交互作用是影响干涉量水平的主要因素。对比模拟值与回归模型预测值发现两者变化趋势一致,且误差不超过10%,表明干涉量多元回归预测模型具有有效性。

This study proposes a method for predicting the installation interference of headless rivet using a multiple nonlinear regression model based on the finite element simulation data of the riveting process,with the aim to elucidate the influence of material properties on the interference during installation and to provide reliable predictions for the use of new materials.Firstly,a finite element simulation model is established based on the actual riveting process and its validity is verified through riveting experiments.Subsequently,finite element analysis and orthogonal testing techniques are employed to examine the significance of various factors,such as elastic modulus,yield strength,strengthening coefficient,and strain strength index,along with their interactions on installation interference,determining the influence of each factor on the riveting interference.Finally,the multiple nonlinear regression model is formulated using a power function,and less significant factors are eliminated.The results show that,under specific riveting process conditions,yield strength and strain strength index,as well as their interaction,are the primary factors affecting the interference.By comparing the simulated values and the predicted values of the regression model,it is observed that the two values exhibit consistent variation trends,with errors remaining below 10%,demonstrating the effectiveness of the multiple regression models in predicting riveting interference.