基于PSO-BP模型的差速器装配密封质量预测  

Research on Sealing Quality Prediction of Differential Assembly Based on PSO-BP model

作　　者：徐静 杨德岭 XU Jing;YANG Deling(College of Mechanical and Electrical Engineering,Northeast Forestry University,Harbin 150040,China)

机构地区：[1]东北林业大学机电工程学院,哈尔滨150040

出　　处：《森林工程》2024年第5期134-144,共11页Forest Engineering

基　　金：黑龙江省应用技术开发与研究项目(GA19C006)。

摘　　要：为了对林业运材车差速器总成装配密封质量进行事前预测,提高其产品质量及装配合格率,提出一种灰色关联分析算法结合粒子群(PSO)优化BP神经网络的预测模型。将由灰色关联分析算法筛选出影响差速器总成密封质量的关键装配工艺参数作为输入变量,差速器总成泄漏值作为输出变量,创建基于粒子群(PSO)算法优化BP神经网络(PSO-BP)的预测模型,结果表明,由灰色关联分析简化后的PSO-BP预测方法得到的平均相对误差最小为1.18%。在此基础上,应用PyQt5 GUI库开发差速器总成泄漏值预测系统。研究结果可以为差速器总成密封质量预测提供理论依据。In order to predict the sealing quality of the differential assembly of forestry timber tranpsort vehicle trucks beforehand and improve the quality of its products and the assembly qualification rate,a prediction model based on grey correlation analysis algorithm combined with particle swarm(PSO)optimized BP neural network is proposed.The key assembly process parameters affecting the sealing quality of differential assembly screened out by the grey correlation analysis algorithm are taken as input variables,and the leakage value of differential assembly is taken as output variable to create a prediction model based on particle swarm algorithm optimized BP neural network,and the results show that the PSO-BP prediction method simplified by the grey correlation analysis obtains the smallest average relative error of 1.18%.On this basis,PyQt5 GUI library is applied to develop a differential assembly leakage value prediction system.The results of the study can provide a theoretical basis for the prediction of differential assembly sealing quality.

关 键 词：运材车辆 差速器 密封质量 灰色关联分析算法 粒子群优化算法 反向传播神经网络 

分 类 号：TH161.7[机械工程—机械制造及自动化]