ArticlePublisher preview available

Spatial weighted graph-driven fault diagnosis of complex process industry considering technological process flow

IOP Publishing
Measurement Science and Technology
Authors:
Fengyuan Zhang at Huazhong University of Science and Technology
Fengyuan Zhang
Jie Liu at Huazhong University of Science and Technology
Jie Liu
Xiang Lu
Xiang Lu
Xiang Lu
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Tao Li
Tao Li
Tao Li
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Show all 8 authorsHide
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Each chemical process industry system possesses unique process knowledge, which serves as a representation of the system's state. As graph-theory based methods are capable of embedding process knowledge, they have become increasingly crucial in the field of process industry diagnosis. The fault representation ability of the diagnosis model is directly associated with the quality of the graph. Unfortunately, simple fully connected graphs fail to strengthen the internal connections within the same process but weaken the interactive connections between different processes. Moreover, each node in the graph is considered equally important, making it impossible to prioritize crucial system monitoring indicators. To address the above shortcomings, this paper presents a spatial weighted graph-driven fault diagnosis method of complex process industry considering technological process flow. Initially, the physical space sensor layout of the technological process flow is mapped into the spatial graph structure, where each sensor is regarded as a node and these nodes are connected by the k nearest neighbor algorithm. Subsequently, according to the mechanism knowledge, the sensors in the process are divided into different importance categories and weight coefficients are assigned to their nodes. The similarities between these weighted nodes are calculated, and the resulting edge information are used to construct the spatial weighted graphs. Finally, the spatial weighted graphs are input to a graph convolutional network, facilitating fault representation learning for fault diagnosis of complex process industry. Validation experiments are conducted using public industrial datasets, and the results demonstrate that the proposed method can effectively integrate the process knowledge to improve the fault diagnosis accuracy of the model.
Figures - available from: Measurement Science and Technology
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by IOP Publishing.
Learn more
Content available from Measurement Science and Technology 
This content is subject to copyright. Terms and conditions apply.
Measurement Science and Technology
Meas. Sci. Technol. 34 (2023) 125143 (17pp) https://doi.org/10.1088/1361-6501/acf665
Spatial weighted graph-driven fault
diagnosis of complex process industry
considering technological process flow
Fengyuan Zhang1, Jie Liu1,∗, Xiang Lu2, Tao Li2,3, Yi Li4,∗, Yingwei Liu4, Lei Tang4
and Hu Wang5
1School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology, Wuhan
430074, People’s Republic of China
2School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology,
Wuhan 430074, People’s Republic of China
3Hubei Three Gorges Laboratory, Yichang 443000, People’s Republic of China
4COFCO (Jilin) Bio-Chemical Technology Co., Ltd, Changchun 130033, People’s Republic of China
5COFCO (Anhui) Bio-Chemical Technology Co., Ltd, Suzhou 234000, People’s Republic of China
E-mail: jie_liu@hust.edu.cn and li-yi1@cofco.com
Received 15 March 2023, revised 24 August 2023
Accepted for publication 1 September 2023
Published 14 September 2023
Abstract
Each chemical process industry system possesses unique process knowledge, which serves as a
representation of the system’s state. As graph-theory based methods are capable of embedding
process knowledge, they have become increasingly crucial in the eld of process industry
diagnosis. The fault representation ability of the diagnosis model is directly associated with the
quality of the graph. Unfortunately, simple fully connected graphs fail to strengthen the internal
connections within the same process but weaken the interactive connections between different
processes. Moreover, each node in the graph is considered equally important, making it
impossible to prioritize crucial system monitoring indicators. To address the above
shortcomings, this paper presents a spatial weighted graph (SWG)-driven fault diagnosis
method of complex process industry considering technological process ow. Initially, the
physical space sensor layout of the technological process ow is mapped into the spatial graph
structure, where each sensor is regarded as a node and these nodes are connected by the k
nearest neighbor algorithm. Subsequently, according to the mechanism knowledge, the sensors
in the process are divided into different importance categories and weight coefcients are
assigned to their nodes. The similarities between these weighted nodes are calculated, and the
resulting edge information are used to construct the SWGs. Finally, the SWGs are input to a
graph convolutional network, facilitating fault representation learning for fault diagnosis of
complex process industry. Validation experiments are conducted using public industrial
datasets, and the results demonstrate that the proposed method can effectively integrate the
process knowledge to improve the fault diagnosis accuracy of the model.
Keywords: complex process industry, technological process ow, sensor layout, fault diagnosis,
weighted graph, graph convolutional network
(Some gures may appear in colour only in the online journal)
∗Authors to whom any correspondence should be addressed.
1361-6501/23/125143+17$33.00 Printed in the UK 1 © 2023 IOP Publishing Ltd
... Statistic learning methods contain PCA [5], LDA [35], and PCA + LDA [36]. Classifcation methods based on deep learning contain CNN [8] and standard GCN [37]. Te details of the proposed MCGFF model are shown in Table 2, and the original learning rate was set to 0.01. ...
Graph Feature Fusion-Driven Fault Diagnosis of Complex Process Industrial System Based on Multivariate Heterogeneous Data
Article
Full-text available
  • Mar 2024
The stable operation of the process industrial system, which is integrated with various complex equipment, is the premise of production, which requires the condition monitoring and diagnosis of the system. Recently, the continuous development of deep learning (DL) has promoted the research of intelligent diagnosis in process industry systems, and the sensor system layout has provided sufficient data foundation for this task. However, these DL-driven approaches have had some shortcomings: (1) the output signals of heterogeneous sensing systems existing in process industry systems are often high-dimensional coupled and (2) the fault diagnosis model built from pure data lacks systematic process knowledge, resulting in inaccurate fitting. To solve these problems, a graph feature fusion-driven fault diagnosis of complex process industry systems is proposed in this paper. First, according to the system’s prior knowledge and data characteristics, the original multisource heterogeneous data are divided into two categories. On this basis, the two kinds of data are converted to physical space graphs (PSG) and process knowledge graphs (PKG), respectively, according to the physical space layout and reaction mechanism of the system. Second, the node features and system spatial features of the subgraphs are extracted by the graph convolutional neural network at the same time, and the fault representation information of the subgraph is mined. Finally, the attention mechanism is used to fuse the learned subgraph features getting the global-graph representation for fault diagnosis. Two publicly available process chemistry datasets validate the effectiveness of the proposed method.
View
View
View
View
Fault diagnosis of rotating machinery based on graph weighted reinforcement networks under small samples and strong noise
Article
  • Mar 2023
  • MECH SYST SIGNAL PR
Available fault vibration signals of large rotating machines are usually limited and consist of strong noise. Existing deep learning methods do not sufficiently extract the correlation relationship between samples, and the process of extracting features is easily disturbed by noise. Therefore, the accuracy of rotating machinery fault diagnosis needs further improvement. A graph-weighted reinforcement network (GWRNet) is proposed to accurately diagnose the faults of rotating machines under small samples and strong noise. First, an adjacency matrix was constructed by measuring the Euclidean distance of the time- and frequency-domain characteristics of small samples to achieve the pre-classification of nodes. Second, the node feature aggregation strategy was designed by dynamically enhancing the largest weight of the multiheaded attention matrix to suppress strong noise interference. Finally, the effectiveness of the proposed method was verified using datasets from the drivetrain diagnostics simulator (DDS) test rig and wind turbine gearboxes.
View
Construction and Evolution of Fault Diagnosis Knowledge Graph in Industrial Process
Article
  • Jan 2022
The steel industry production line is complicated and contains a substantial amount of equipment, leading to serious problems in fault diagnosis such as wrong inspection strategies, fault location, maintenance, and so on. To realize accurate and efficient equipment fault diagnosis, this paper proposes a steel production line equipment fault diagnosis knowledge graph (SPLEFD-KG) based on a novel relation-oriented model with global context information for jointly extracting overlapping relations and entities (ROMGCJE). A low-level self-learning SPLEFD-KG is first constructed using the triples extracted by ROMGCJE. However, this low-level SPLEFD-KG is incomplete, and only contains sparse paths for fault reasoning. To overcome this problem, a reinforcement learning framework is applied to mine hidden semantic knowledge to complete the missing relation. Besides, the graph neural networks are introduced to compute the embedding vector of new entities outside of the SPLEFD-KG for continuously completing missing entities. Finally, the low-level self-learning SPLEFD-KG evolves to one high-level SPLEFD-KG, which can provide information-rich and accurate fault-related knowledge. Extensive experiments conducted on the steel production line equipment failure dataset indicate that the novel SPLEFD-KG significantly improves fault diagnosis results and provides effective maintenance programs.
View
Global-Local Integration for GNN-based Anomalous Device State Detection in Industrial Control Systems
Article
  • Aug 2022
  • EXPERT SYST APPL
Anomaly detection are gaining popularity among the research communities for its essential role in securing Industrial Control Systems (ICS). Over the decades, diverse approaches have been proposed to profile anomalous behaviours propagating across the ICS networks. Recent attempts using the Graph Neural Network (GNN) methodologies have enabled state prediction of a device node via encoding its immediate neighbourhood. Such an encoding scheme potentially compromises the model’s detection accuracy due to the nodes’ biased attention towards their local surroundings. To investigate this issue, we present the Global-Local Integration Network (GLIN) that achieves node-level anomaly detection by merging a node’s local and the network’s global expressions. It comprises a preprocessor for graph construction and data transformation, an encoder for node embedding learning, a pooling module producing global representations, an integration module that performs message fusion, and a decoder for label prediction. We develop and evaluate GLIN with 7 global integration schemes and train it over 3 message passing mechanisms. We compare its performance against both classical machine learning and recent deep learning baselines and demonstrate its superiority in terms of multiple popular metrics. Finally, we provide useful insights on the results and suggest promising future work directions.
View
Pruning graph convolutional network-based feature learning for fault diagnosis of industrial processes
Article
  • May 2022
  • J PROCESS CONTR
In recent years, deep learning has been widely applied in process fault diagnosis due to its powerful feature extraction ability. A predominant property of these fault diagnosis models is to extract effective features from process signal. However, it is still difficult for them to construct the feature association relationship between input data. To solve these problems, this paper proposes a new graph neural network (GNN), pruning graph Convolutional network (PGCN), to perform feature learning based on the graph data. One dimensional process data are transformed into graph data by a graph construction method. A graph Convolutional network (GCN) is used to extract the features of process data. A pruning method of graph structure is proposed to effectively extract important information from process fault data. The feasibility and effectiveness of PGCN are verified on two benchmark processes, i.e., continuous stirred-tank reactor (CSTR) and fed-batch fermentation penicillin process (FBFP). The experimental results show that the performance of PGCN in feature extraction and process fault diagnosis is better than that of other typical methods, which provides a good possibility for the application of GCN in industrial process fault diagnosis.
View
Data-driven identification model for associated fault propagation path
Article
  • Dec 2021
  • MEASUREMENT
In this paper, a data-driven identification model of the associated fault propagation path is proposed. Different from traditional fault diagnosis methods, the alternative approach is focused more on identifying the fault propagation path. Firstly, the KPCA method is used for fault detection. Then, a new transfer entropy method is proposed to construct the causality diagram. Finally, a kernel extreme learning machine-based search approach is proposed for fault propagation path identification. To demonstrate the effectiveness and applicability, the proposed model is applied to the Tennessee-Eastman Process and the real telemetry data of the satellite in orbit. Experimental results show that the number of causal edges obtained by the proposed method is only 12.62% of the traditional transfer entropy method, and 24.45% of the multivariate transfer entropy method.
View
Process Topology Convolutional Network Model for Chemical Process Fault Diagnosis
Article
  • Apr 2021
  • PROCESS SAF ENVIRON
There always exists potential safety risk in chemical processes. Abnormalities or faults of the processes can lead to severe accidents with unexpected loss of life and property. Early and accurate fault detection and diagnosis (FDD) is essential to prevent these accidents. Many data-driven FDD models have been developed to identify process faults. However, most of the models are black-box models with poor explainability. In this paper, a process topology convolutional network (PTCN) model is proposed for fault diagnosis of complex chemical processes. Experiments on the benchmark Tennessee Eastman process showed that PTCN improved the fault diagnosis accuracy with simpler network structure and less reliance on the amount of training data and computation resources. In the meantime, the model building process becomes much more rational and the model itself is much more understandable.
View
Intelligent fault diagnosis for rolling bearings based on graph shift regularization with directed graphs
Article
  • Feb 2021
  • ADV ENG INFORM
Graph shift regularization is a new and effective graph-based semi-supervised classification method, but its performance is closely related to the representation graphs. Since directed graphs can convey more information about the relationship between vertices than undirected graphs, an intelligent method called graph shift regularization with directed graphs (GSR-D) is presented for fault diagnosis of rolling bearings. For greatly improving the diagnosis performance of GSR-D, a directed and weighted k-nearest neighbor graph is first constructed by treating each sample (i.e., each vibration signal segment) as a vertex, in which the similarity between samples is measured by cosine distance instead of the commonly used Euclidean distance, and the edge weights are also defined by cosine distance instead of the commonly used heat kernel. Then, the labels of samples are considered as the graph signals indexed by the vertices of the representation graph. Finally, the states of unlabeled samples are predicted by finding a graph signal that has minimal total variation and satisfies the constraint given by labeled samples as much as possible. Experimental results indicate that GSR-D is better and more stable than the standard convolutional neural network and support vector machine in rolling bearing fault diagnosis, and GSR-D only has two tuning parameters with certain robustness.
View
View
Vibration fault features of planetary gear train with cracks under time-varying flexible transfer functions
Article
  • Apr 2021
  • MECH MACH THEORY
Due to the multiple gear-pair meshes in planetary gear train, fault features of tooth root crack are weak and complicated in vibration signals. Influenced by the time-varying transfer paths of vibration signals, the fault features of crack are further weakened, which becomes hard to diagnose. To reveal the origination of crack fault features, a lumped-parameter-model of planetary gear train is firstly built to solve the excitation forces that generate vibration signals. Gear mesh stiffness that drives the excitation forces is used to equivalently reflect the crack characteristics. Secondly, a finite-element-model of flexible housing is used to calculate the time-varying transfer functions that determine the amplitude characteristics of vibration signals. Based on the signal convolution theorem, the full-band vibration signal is deduced by the convolution of excitation forces and flexible transfer functions. Simulations and experimental tests on two typical types of planetary gear trains show good consistency at crack fault frequency features, which effectively validates the proposed models. Significant fault features locating at resonance modulation sidebands are summarized as effective indicators for crack diagnosis.
View
Sparse dictionary design based on edited cepstrum and its application in rolling bearing fault diagnosis
Article
  • Sep 2020
  • J SOUND VIB
Rolling bearing with a localized defect usually generates periodically impact vibration responses, which carry important information for bearing fault diagnosis. Due to the inevitable noise disturbances, extracting accurate impact features of faulty bearing is still a hard task. In view of the superiority of sparse decomposition on feature extraction, a novel sparse dictionary design method is proposed based on edited cepstrum to improve the precision of feature extraction. The impulse response function is selected as sparse atom, which better reflects the structure and inherent modal characteristics of the faulty bearing. The modal parameters are directly identified from the deconvolved fault signal by edited cepstrum. Identification errors caused by the cepstrum windowing are corrected by quantitative compensation, which further improves the accuracy of dictionary design. A segmental matching pursuit algorithm is applied to speed sparse coefficients solving and fault features reconstruction. A series of simulation analyses verify the proposed method's effectiveness, anti-noise performance and robustness. Experimental tests on pure rolling bearing and gearbox bearing further verify the method's effectiveness under different working conditions. Additionally, comparisons with an improved spectral kurtosis method and an edited cepstrum methodshow the proposed method be more reliable in diagnostic performance.
View