摘要
针对工业过程故障诊断中数据规模的逐渐增大以及数据之间映射关系、复杂程度的增加,文中以TE过程为数据背景提出一种基于灰狼算法(Grey Wolf Optimizer,GWO)优化卷积神经网络(GWO Convolutional Neural Networks,GWO-CNN)的模型,结合GWO算法具有搜索能力强、结构清晰、容易实现等特点,寻找CNN卷积核的个数等参数的最优解,并利用所寻得的最优参数搭建GWO-CNN模型并将其应用于工业过程的故障诊断。仿真结果表明,相比传统的卷积神经网络,GWO-CNN算法能够从原始数据中提取更多故障特征,从而提升原有的故障诊断的准确率。
Based on the fact that the gradual increase in the scale of data and the increase in the mapping relationship and complexity between data in the process of industrial fault diagnosis,this paper proposes a model based on Grey Wolf Optimizer(GWO)to optimize the convolution neural network(GWO-CNN)with the TE process as the data background.Combined with the characteristics of GWO algorithm such as strong search ability,clear structure and easy realization,the optimal solution of parameters such as the number of convolution kernels of CNN is found.The GWO-CNN model is built with the obtained optimal parameters and applied to the fault diagnosis of industrial processes.The simulation results show that compared with the traditional convolutional neural network,GWO-CNN algorithm can extract more fault features from the original data,so as to improve the original fault accuracy.
作者
赵芷锐
李元
ZHAO Zhi-rui;LI Yuan(School of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China)
出处
《信息技术》
2024年第7期121-127,共7页
Information Technology
基金
国家自然科学基金项目(61673279)。
关键词
灰狼优化算法
卷积神经网络
田纳西-伊斯曼过程
故障诊断
机器学习
Grey Wolf Optimization
Convolutional Neural Network
Tennessee Eastman process
fault diagnosis
machine learning
