基于MPSO-BP对5A06铝合金薄壁件变形预测

为预测不同铣削参数下的5A06铝合金薄壁件的加工变形,文章基于BP神经网络和粒子群算法提出了一种新的方法,该方法对粒子群算法中的惯性权重和学习因子进行动态调整并提出了新的惯性权值自适应策略,之后对相关参数进行优化形成改进粒子群优化算法,最后用改进后的粒子群算法优化BP神经网络并将优化后的BP神经网络用于5A06铝合金薄壁件加工变形预测。仿真实验结果表明:MPSO-BP相对于PSO-BP和BP有较小的预测误差,现场加工实验结果进一步说明了MPSO-BP具有良好的预测精度。

The Deformation Prediction of 5A06 Aluminum Alloy Thin-wall Parts Based on MPSO-BP

In order to predict the machining deformation of 5 A06 aluminum alloy thin-walled parts under different milling parameters, the paper proposes a new method based on BP neural network and particle swarm optimization algorithm, which dynamically adjusts the inertia weight and learning factor in particle swarm optimization,what is more, a new inertia weight adaptive strategy is proposed. Then, the relevant parameters are optimized to form an improved particle swarm optimization algorithm. In the last, the improved particle swarm optimization algorithm is used to optimize BP neural network and the optimized BP neural network is used for 5 A06 aluminum alloy thin-wall piece machining deformation prediction. The simulation results show that MPSO-BP has less prediction error than PSO-BP and BP and the results of field processing experiments further demonstrate that MPSO-BP has good prediction accuracy.