基于KPCA-PIO-ELM模型的管道剩余寿命预测分析  

Prediction and analysis of pipeline remaining life based on KPCA-PIO-ELM model

作　　者：霍奕宇 李西锋 HUO Yi-yu;LI Xi-feng(Shaanxi Institute of Technology,Xi’an 710300,China)

机构地区：[1]陕西国防工业职业技术学院机械工程学院,西安710300

出　　处：《信息技术》2024年第10期88-93,共6页Information Technology

基　　金：陕西国防工业职业技术学院科研项目(Gfy22-63)。

摘　　要：为了提高腐蚀管道剩余寿命预测精度,提出基于核主成分分析(KPCA)和鸽群优化算法(PIO)的极限学习机(ELM)预测模型。通过KPCA提取关键腐蚀因素,降低预测指标维度;采用PIO对ELM的输入权值及隐层阈值进行优化,提升预测精度。为检验模型效能,以某注水管道的50组数据为例进行研究,并与ELM、BP两组模型的预测结果进行对比分析,结果表明:构建模型的MAE、MAPE、RMSE均优于对比模型,证明KPCA-PIO-ELM模型在预测注水管道剩余寿命方面具有可行性及优越性。In order to improve the prediction accuracy of remaining life of corroded pipelines,an Extreme Learning Machine(ELM)prediction model based on Kernel Principal Component Analysis(KPCA)and Pigeon Colony Optimization algorithm(PIO)is proposed.The key corrosion factors are extracted by KPCA to reduce the dimension of prediction index.PIO is used to optimize the input weight and hidden layer threshold of ELM to improve the prediction accuracy.In order to test the efficiency of the model,50 sets of data of a water injection pipeline are taken as an example to study,and compared with ELM and BP models.The results show that MAE,MAPE and RMSE of the model are better than the comparison model,which proves that KPCA-PIO-ELM model is feasible and obviously superior in predicting the remaining life of water injection pipeline.

关 键 词：剩余寿命预测 腐蚀管道 核主成分分析 鸽群优化算法 极限学习机 

分 类 号：TG178[金属学及工艺—金属表面处理]