检索结果-内蒙古大学图书馆

Early identification of process deviation based on convolutional neural network

Chinese Journal of Chemical Engineering 2023年第4期56卷 104-118页

作者： Fangyuan Ma Cheng Ji Jingde Wang Wei Sun College of Chemical Engineering Beijing University of Chemical TechnologyBeijing 100029China Center of Process Monitoring and Data Analysis Wuxi Research Institute of Applied TechnologiesTsinghua UniversityWuxi 214072China

A novel process monitoring method based on convolutional neural network(CNN)is proposed and applied to detect faults in industrial *** utilizing the CNN algorithm,cross-correlation and autocorrelation among variables are captured to establish a prediction model for each process variable to approximate the first-principle of physical/chemical relationships among different variables under normal operating *** the process is operated under pre-set operating conditions,prediction residuals can be assumed as noise if a proper model is *** process faults occur,the residuals will increase due to the changes of correlation among variables.A principal component analysis(PCA)model based on the residuals is established to realize process *** monitoring the changes in main feature of prediction residuals,the faults can be promptly *** studies on a numerical nonlinear example and data from two industrial processes are presented to validate the performance of process monitoring based on CNN.

关键词： process monitoring Residual Principal component analysis process systems Systems engineering

来源：评论

学校读者我要写书评

暂无评论

Fault Detection for Industrial Chemical Production Using Siamese Autoencoder

引用

IFAC-PapersOnLine 2024年第14期58卷 823-828页

作者： Cheng Ji Fangyuan Ma Jingde Wang Wei Sun College of Chemical Engineering Beijing University of Chemical Technology 100029 North Third Ring Road 15 Chaoyang District Beijing China Center of process monitoring and data analysis Wuxi Research Institute of Applied Technologies Tsinghua University 214072 Wuxi China

Nowadays, deep learning has emerged as a transformative technology in various domains, including process monitoring. Massive advanced deep learning algorithms, such as autoencoder, recurrent neural network, and convolutional neural network, have been explored in the application of chemical processes to enhance the overall monitoring performance. Nevertheless, deep learning models often imply complex structures and a huge number of parameters, leading to limited generalization ability when they are employed in industrial chemical processes. In this work, the above limitation is addressed by monitoring the representative and discriminative features extracted in the latent space by Siamese autoencoder. The reconstruction ability and discriminative information between a pair of inputs are considered in the extraction of latent features, by which better process monitoring performance can be achieved with much fewer model parameters. Case study on an industrial chemical process is investigated to demonstrate the effectiveness of the proposed method.

关键词： process monitoring deep learning Siamese neural network catalytic reforming heat exchanger

来源：评论

学校读者我要写书评

暂无评论

Molecular Reconstruction of Naphtha based on Physical Information Neural Network

引用

IFAC-PapersOnLine 2022年第7期55卷 186-191页

作者： Fangyuan Ma Xin Zheng Chengyu Han Jingde Wang Wei Sun College of Chemical Engineering Beijing University of Chemical Technology 100029 Beijing China Center of process monitoring and data analysis Wuxi Research Institute of Applied Technologies Tsinghua University 214072 Wuxi China

A molecular reconstruction method based on physical information neural network is proposed for predicting the molecular composition of naphtha. By embedding physical information utilized in typical molecular reconstruction methods, such as mixing rules, into the loss function of the neural network, the model tends to converge to the state conforming to physical rules in training stage. The neural network model obtained by the method contains certain physical information, which can improve the generalization ability of the model. The results show that the prediction performance and application range of the proposed method are better than those of the typical ANN-based molecular reconstruction method.

关键词： Artificial Neural Network Mixing Rules Generalization Ability

来源：评论

学校读者我要写书评

暂无评论

A cyber-attack detection based on time-delay mutual information analysis

引用

process Safety and Environmental Protection 2025年 200卷

作者： Fangyuan Ma Cheng Ji Jingzhi Rao Chengyu Han Jingde Wang Wei Sun College of Chemical Engineering Beijing University of Chemical Technology Beijing 100029 China Center of Process Monitoring and Data Analysis Wuxi Research Institute of Applied Technologies Tsinghua University Wuxi 214072 China

In recent years, data-driven fault detection and diagnosis (FDD) have been widely studied and applied in academia and industry to ensure the safe operation of processes. However, traditional FDD is mainly developed to process spontaneous faults and ignores the detection of cyber-attack. With the widespread application of industrial control systems, a large number of control actuators and measurement sensors are connected using wireless or wired communications, which greatly increases the risk of sensor tampering. Meanwhile, faults caused by sensor tampering have similar effects as process faults, making the distinction much more difficult. If the fault is incorrectly recognized as a process fault, it will mislead human engineers, even cause catastrophic consequence. Therefore, it is important to identify cyber-attacks promptly and accurately. By comparing the two types of faults, the main difference lies in the root cause of the fault. In a cyber-attack, state variables such as temperature, pressure, level, etc. are the preferred targets to be tampered with, and feedback from industrial control systems is not able to control them to a preset value. Therefore, the root cause of these faults is reflected in state variables. In contrast, changes in the manipulated variables, such as feed flow, are frequently the root cause of process faults. Inspired by it, the two types of faults should be distinguished by the root cause. Time-delay mutual information analysis (TDMI) is a fault diagnosis method that has been successfully applied to actual industrial processes. By analyzing the changes in time-delay mutual information between variables, the fault propagation path and the root cause can be objectively identified. In this work, a cyber-attack detection based on TDMI is developed. When a fault occurs, TDMI is employed to analyze the root cause. If the root cause is a state variable, the fault should be caused by a cyber-attack in a big chance, instead of a process fault.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Differential recurrent autoencoder for early fault detection of nonstationary industrial processes

引用

Engineering Applications of Artificial Intelligence 2025年 156卷

作者： Ji, Cheng Tao, Tingting Ma, Fangyuan Wang, Jingde Sun, Wei College of Chemical Engineering Beijing University of Chemical Technology North Third Ring Road 15 Chaoyang District Beijing100029 China Center of Process Monitoring and Data Analysis Wuxi Research Institute of Applied Technologies Tsinghua University Wuxi214072 China

Artificial Intelligence (AI) has emerged as a prominent technique for industrial fault detection, where deep learning demonstrates unique advantages in modeling nonlinear and dynamic process systems. Nevertheless, the presence of stochastic factors and permissible external disturbances frequently induce nonstationary characteristic during industrial process operation, which poses significant obstacles to early fault detection, particularly in capturing time-varying statistical patterns that precede critical failures. To bridge this gap, we propose a differential recurrent autoencoder (DRAE) for early fault detection of nonstationary industrial processes. DRAE based on differential operation, aiming to optimize the update mechanism of the forget gate to enhances its capacity to capture short-term nonstationary features. To facilitate the transition of time-variant nonstationary information in memory states, differential operations are embedded into the latent layer of Long Short-Term Memory (LSTM) neural network. The integration of the differential gradient layer with LSTM units directly models transient nonstationary dynamics while retaining original sequential dependencies, thereby overcoming the stationarity assumption inherent in conventional process monitoring approaches. Combined with an unsupervised encoder-decoder structure, this framework enables automatic feature extraction through fault-discriminative reconstruction to perform early fault detection. The proposed method is validated on a nonstationary numerical process, the Tennessee Eastman (TE) process, and a real-world chemical industrial process. The experimental results demonstrate significant improvement in fault detection rate and reduced detection delays compared to baseline methods, showcasing the potential of AI in enhancing process safety management within industrial systems. © 2025 Elsevier Ltd

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：