基于PSO-BP神经网络的小冲孔蠕变寿命预测模型  

Prediction Model of Small Punch Creep Life Based on PSO-BP Neural Network

作　　者：王杰[1] 郑杨艳 凌祥[1] WANG Jie;ZHENG Yang-yan;LING Xiang(School of Mechanical and Power Engineering;School of Energy Science and Engineering,Nanjing Tech University)

机构地区：[1]南京工业大学机械与动力工程学院 [2]南京工业大学能源科学与工程学院

出　　处：《化工机械》2023年第3期348-353,387,共7页Chemical Engineering & Machinery

摘　　要：为提高小冲孔蠕变寿命预测的准确性,选取温度、载荷参数作为预测模型的输入参数,利用粒子群算法(PSO)优化神经网络权值和阈值,建立基于PSO-BP神经网络的小冲孔蠕变寿命预测模型。使用MATLAB建立优化后的预测模型,并与传统BP神经网络预测的结果进行对比。结果表明:粒子群优化的神经网络有效提高了预测模型的准确性和稳定性。可见,PSO-BP小冲孔蠕变寿命预测模型是合理、可行的,为小冲孔蠕变寿命预测提供了一种简单可行的思路和办法。For purpose of improving the creep life prediction accuracy of small punches,having the temperature and load parameters selected as the input parameters of the prediction model was implemented,and the neural network weights and thresholds were optimized by using the particle swarm algorithm(PSO)to establish a small punch creep life prediction model based on the PSO-BP neural network.The optimized prediction model was built through using MATLAB and compared with the results predicted by the conventional BP neural network.The results show that,the neural network optimized by particle swarm effectively improves the accuracy and stability of the prediction model.It can be seen that the PSO-BP small punch creep life prediction model is reasonable and feasible to provide a simple and feasible idea and approach for small punch creep life prediction.

关 键 词：小冲孔试验 蠕变 PSO-BP神经网络 寿命预测 

分 类 号：TG113.25[金属学及工艺—物理冶金]