摘要
针对压铸成型的工艺参数系统难以建立精确的数学模型,参数优化凭经验试凑,难以得到最优成型工艺参数的问题。本文提出了一种基于量子粒子群算法改进的Kriging算法来建立压铸成型工艺参数系统精确的数学模型,通过Kriging代理模型技术建立工艺参数与控制量之间的精确数学模型,采用量子粒子群算法对Kriging代理模型的变差函数的参数进行优化,提高KRIGING建立的工艺参数与控制量之间的数学模型精度。仿真结果表明:基于量子粒子群算法改进后的Kriging模型精度评价指标的R2提高了9.4741%,RMSE降低了82.3207%,RMAE降低了84.9139%,预测误差更小,由原来的[-2,10]优化为[-2,1]之间,提高了Kriging模型的拟合精度。
Aiming at the issue that it is difficult to establish an accurate mathematical model for the technical parameter system of die casting,parameter optimization is realized on the basis of experience,and therefore,it is difficult to find the optimal process parameters.This paper proposes an improved Kriging algorithm based on quantum particle swarm optimization(QPSO)with levy in order to establish an accurate mathematical model for the process parameter system of die casting,The mathematical model between process parameters and control quantities was established by Kriging model technology,and the parameters of variation function of Kriging model were optimized by using the quantum particle swarm optimization algorithm to improve the accuracy of the mathematical model between process parameters and control quantities established by Kriging,The simulation results show that the R2 of the improved Kriging model is improved by 9.4741%,RMSE by 82.3207%,and RMAE by 84.9139%.The prediction error is smaller and the original[-2,10]is optimized to[-2,1],which improves the fitting accuracy of the Kriging model.
作者
赵瑜
ZHAO Yu(Jiangsu Vocational College of Information Technology,Wuxi 214153,Jiangsu)
出处
《攀枝花学院学报》
2020年第5期42-46,共5页
Journal of Panzhihua University
