Operation Of Industrial Processes Research Articles

Fault diagnosis of complex industrial equipment based on chunked statistical global-local feature fusion

Abstract Industrial process equipment is bulky and complex in structure, which is easy to produce faults during operation and affect production efficiency, cause huge economic losses, and even threaten the safety of workers. To achieve sustainable operation of large-scale industrial processes, timely and accurate monitoring and handling of abnormal situations are essential. However, fault monitoring of large equipment requires the collection of abundant data, which includes many complex related variables, resulting in excessive redundant data generated during the fault monitoring process. Moreover, the existing principal component analysis (PCA) method can only retain the global characteristic of variance information, and cannot obtain the local characteristic that can characterize the topological relationship between the data points, which affects its monitoring reliability and intelligence level. In response to these issues, a fault diagnosis model for complex industrial processes based on chunked statistical global-local feature fusion (CSGLFF) is proposed in this paper. First, considering the correlation characteristics between industrial process variables, a correlation variable chunking method mutual information-based is designed to merge the variables with small correlation to obtain the optimal chunking of variables. Second, PCA and locality preserving projections (LPP) are combined to construct a global-local feature fusion (GLFF) model that can extract global and local features simultaneously. The chunked data are imported into the GLFF for the extraction of its features respectively, and the corresponding CSGLFF is established. In addition, Bayesian inference is used to fuse the statistics of each sub-chunk to establish an overall fault monitoring statistical indicators, and the reason for failure is found through the variable contribution graph. Finally, two cases of Tennessee Eastman process (TEP) and laboratory emulsion pump were used to conduct experimental research on the performance of CSGLFF. The results show that compared with the chunked statistical PCA, chunked statistical LPP, and GLFF algorithms, The accuracy of fault monitoring for TEP mean, flow pulsation impact, and pressure anomaly of this method reached 92.91%, 97%, and 90.30%, respectively. It has good monitoring effect in processing data with large variables, reducing the generation of redundant data, improving the accuracy of industrial monitoring, and accurately identifying the relevant variables of fault occurrence. This provides a theoretical basis for determining the fault location and points out the direction for maintenance by staff.

Optimized structure design for binary particle mixing in rotating drums using a combined DEM and gaussian process-based model

Particle mixing is a crucial operation in various industrial production processes. However, phenomena like segregation or local accumulation can arise, especially when particles differ in properties like radius and density. Numerical simulation of particles using Discrete Element Method (DEM) allows for the manipulation of control variables in batches, generating a large amount of data and facilitating quantitative research. In this study, the mixing behaviors of binary particles in rotating drums are systematically investigated. The DEM model is first validated with experimental data and then rotating drums with varying obstacles, rotation speeds, particle radii, and densities are simulated. Moreover, a Gaussian process-based optimization is conducted by correlating Lacey mixing index (MI) and parameterized shape of obstacle to find the optimized mixing condition. Experimental validations are further performed on the optimized condition to verify the design. It is shown that this integrated approach holds significant potential for enhancing the efficiency, effectiveness of industrial mixing processes and the consideration of energy consumption when balancing the mixing efficiency and optimal rotating speed.

Evaluation of Hazards and Operability Study for Optimization Operation of Industrial Processes

Nitrocellulose is a substance that ignites easily when exposed to a hot or humid environment. It is widely used in the military as well as civil industries, most notably the production of explosives and films. Therefore, it was selected as a case study in order to assess the risks in the seven production stages using HAZOP (Hazards and Operability) technology. In this study, the raw material preparation stage investigated. This stage was divided into five identified nodes. The deviation from the design intention for 16 critical parameters for this process was studied.The findings of the study were as follow: Nitric acid and sulfuric acid are stored for a long time under process shutdown resulting in the corrosion of the tank and acids decomposition, Acids measurement depend on operator competence in the process of preparation of the acid mixture, which leads to inappropriate operating conditions, the temperature of the acid mixture increases due to endothermic mixing process resulting in decomposition of the mixture, in the process of feeding RC to the reactor, the flow increases due to the increase in the speed of the belt affecting the degree of nitration. Recommendations are; It should not be turned off until the end of the acids, thermoneutral process was wanted and Adjust the RC flow speed every 10 operations.

Manifold regularized deep canonical variate analysis with interpretable attribute guidance for three-phase flow process monitoring

Oil–gas–water three-phase flow has multiple flow states, exhibiting dynamic, nonlinear, and instantaneous behaviors. Monitoring and analysis of flow state are crucial for ensuring safe operation of industrial processes. However, the absence of a universal definition for three-phase flow states, owing to the complex flow characteristics and structures, leads to that labeling three-phase flow states by categories is impractical. For comprehensive monitoring and analysis, the flow states should be described from both global (e.g., the dominant phase) and local (e.g., the interphase structure) aspects. Given the aforementioned challenges, the work aims to address the accurate identification of typical three-phase flow states and to meticulously monitor and analyze the continuous evolutionary process of three-phase flow. To achieve it, the manifold regularized deep canonical variate analysis with attribute guidance (Ag-MRDCVA) is proposed, which not only introduces an innovative monitoring paradigm that designs state attributes for process description but also facilitates knowledge transfer from typical flow states to transition states. For enhancing the model’s feature embedding capabilities, an attribute guidance module is introduced to increase supervision information at both class and attribute levels. Then, convolutional neural network (CNN) backbones are developed with a dual objective of global serial correlation and local manifold regularization, which capture spatiotemporal information at both global and local scales, facilitating effective attribute encoding and characterization of flow states. Finally, the attribute evolution heat map and a monitoring metric (MM) collectively offer a compelling and comprehensive analysis of the three-phase flow process. Extensive experiments demonstrate that Ag-MRDCVA surpasses existing methods, showcasing its ability to minutely monitor the process while providing reasonable explanations.

Recent University Research, Implementation and Education in Support of Pyrometallurgical Lead Processing

The increasing chemical complexity of lead process streams encountered in industrial high temperature processing operations, as the result of declining primary resources, increased metal recycling and increased overall range of metals in modern devices has highlighted the urgent need for new predictive tools, fundamental phase equilibria and thermodynamic information and thermodynamic models to characterise the chemical behaviour of these systems. The paper examines recent progress in experimental and thermodynamic modelling research on process fundamentals, the availability of advanced, predictive computer-based tools and the implementation of the research outcomes into industrial practice. A wide range of chemical systems and phase assemblages have been studied. Some examples are taken from the current research program at PYROSEARCH, which involves the characterisation of multi-component, multi-phase gas-slag-matte-speiss-metal-solids systems with the PbO-ZnO-“Cu2O”-FeO-Fe2O3-CaO-Al2O3-MgO-SiO2-S as major and As-Sn-Sb-Bi-Ag-Au-Ni-Co-Cr-Na as minor elements with focus on systems directly relevant to lead primary and recycling pyrometallurgical processes. Examples of the application of advanced analytical techniques to fundamental and applied industrial research are also given. The implementation of new research outcomes into industrial practice depends critically on commitments by research staff as well as industry management and the availability of well-trained metallurgical engineers. We examine the current status of research implementation, university research, metallurgical engineering education and the availability of suitable educational pathways and initiatives that can be taken to increase undergraduate enrolments. Active engagement and support by industry is critical in ensuring the continuation of academic programs and advanced technical skills required by the industry.

Interval Type-2 Fuzzy Neural Network Based on Active Semi-Supervised Learning for Non-Stationary Industrial Processes

Accurate models ensure the efficient and safe operations of industrial processes. However, modeling of industrial processes with non-stationary characteristic is challenging. In this study, an interval type-2 fuzzy neural network (IT2FNN) based on active semi-supervised learning (ASSL-IT2FNN) is proposed for such industrial processes. First, an IT2FNN with adaptive fuzzy membership function (FMF) and hierarchical learning method is utilized to achieve considerable modeling accuracy and efficiency. Second, to better tackle with the non-stationary characteristic, an unsupervised distribution detection method is proposed to identify the occurrence moments of concept drift actively from the perspective of probability and spatial projection. Then, by an active semi-supervised learning method, the concept drift samples with partial labels are used to build a subnetwork, guaranteeing the performance of the constructed IT2FNN models with incomplete data in non-stationary environments. Finally, the proposed ASSL-IT2FNN is verified by benchmark simulations and real industrial data, and the experimental results demonstrate its outperformance. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Concept drift is an important indicator of a non-stationary environment. It manifests as changes in the distribution of samples over time, which may reduce the accuracy of models constructed based on historical samples. This study aims to alleviate the impact of concept drift on model performance in non-stationary industrial processes while reduce the cost of sample labeling. By using the adaptive FMF and hierarchical learning method, the designed IT2FNN can effectively deal with the inherent nonlinearity of complicated industrial processes. The samples representing concept drift are actively identified by the Kullback-Leibler (KL) divergence and maximum mean difference (MMD), enabling timely detections of concept drift moments in unsupervised situations. The active semi-supervised learning is used to adapt the constructed model to handle the concept drift, wherein a subnetwork is designed based on concept drift samples and similar historical samples.

Modeling and Simulation of Hydroxyapatite Recovery in the Desliming Circuit of the Tapira Industrial Plant, Brazil

The modeling and simulation of industrial mineral processing operations are traditionally used for cyclone sizing and optimizations of industrial operations. However, the main models used are based on the total population of particles in the pulp, thus not distinguishing the individual minerals. This article presents the results of an innovative method that investigated the optimization of the metallurgical recovery of P2O5 in the desliming circuit of a phosphate ore processing plant in Brazil. A survey campaign was carried out in the existing industrial circuit, followed by determining the partition curves for the overall particles and specifically for the hydroxyapatite particles. The results were used to calibrate the Narasimha–Mainza cyclone model. From a Base Case determined with reference to the industrial survey, three optimization scenarios were simulated through cyclone geometries and respective operating conditions changes. Simulated scenarios indicated the possibility of P2O5 metallurgical recovery increasing from 9.4% to 12.7% compared to the Base Case.

Joint structure bipartite graph projection and its application for industrial process monitoring

Process monitoring plays a crucial role in maintaining the safety and stable operation of industrial processes. One of its key tasks is to improve the accuracy of detecting abnormal conditions. To this end, we propose a dimensionality reduction method called Joint Structure Bipartite Graph Projection. This method not only captures global information, but also incorporates manifold learning and anchor strategy to capture the local structure of data in an adaptive neighbor manner. In constructing the bipartite graph between projected samples and projected anchors, structural information between samples and anchors is fully considered to ensure an effective graph. Furthermore, we introduce an information entropy-based method to compute the balance parameters, reducing costs of parameter tuning. Ultimately, we develop a process monitoring model based on JSBGP method, achieving the goals of anomaly detection and abnormal variables isolation. Through experiments on benchmark dataset and actual industrial process demonstrate the effectiveness of the proposed method.

Unsupervised feature selection using chronological fitting with Shapley Additive explanation (SHAP) for industrial time-series anomaly detection

With the development of the industrial Internet of Things (IIOT), an amount of industrial multivariate time series (IMTS) data has been collected by various sensors. IMTS data anomaly detection plays an important role in industrial process monitoring and operation condition identification. Many real-world IMTS datasets usually have a large number of redundant features, which may lead to deviation of the anomaly detection model and misunderstanding of potential feature correlation. Hence, it is essential to select features that have critical and effective effects on anomalies and operation conditions. In this paper, a feature selection that introduces knowledge-based target variables into inner feature selectors (IFSs) is proposed as a wrapper method, referred to as ILSFS. The IFS is constructed by integrating a timing fitter, which is established based on the long and short-term memory (LSTM) network, with the Shapley Additive explanation (SHAP) assisted to eliminate unimportant features. The experiments were conducted on two datasets: the vertical roller mill dataset (VMD) of a slag milling factory in Jiangsu Province and the public server machine dataset (SMD). Compared with None feature selection, SVM-RFE, LGBM -Boruta, Genetic-based methods, and other SHAP-assisted feature selection methods such as XGBSVFIR and ShapHT+, the experiment results highlighted the proposed methodology ILSFS improves the best F1 score of each the above baselines by 0.2799, 0.2208, 0.2647, 0.1882, 0.3775, 0.2765 for the VMD and 0.1475, 0.0586, 0.0567, 0.0638, 0.0519, 0.2250 for the SMD on the downstream anomaly detection. The best precision of the ILSFS for the VMD and the SMD was 0.8224 and 0.9499, respectively, while its best recall was 0.7353 and 0.4902, respectively.

An industrial process monitoring method and its application with fractal-based structure preserving embedding

Maintaining stable industrial process operation is a crucial research topic in modern process monitoring. However, industrial process data often contains many redundant correlated variables, which can disrupt the process of data dimensionality reduction. To address this, a novel dimensionality reduction method termed Fractal-based Structural Preservation Embedding (FSPE) is proposed. This introduces a new feature selection approach via fractal analysis, which effectively eliminates redundant variables without altering data orientation, enhancing interpretability. The data structures are then preserved at spatial and temporal scales by solving a dual-objective optimization function. Spatially, it preserves the local data manifold structure while considering global information. Temporally, it maintains time series by searching for nearest neighboring points in time. The efficacy of the FSPE monitoring method is verified through simulations involving the Tennessee Eastman Process (TEP) and the Electric Furnace Magnesia Furnace (EFMF).

Measurement of void fraction in gas-water bubbly flow using a derived multi-eigenvalue sequence from normalized EIT impedance matrix

Void fraction is one of the dominant parameters of gas-water two-phase flow. Its accurate measurement plays an important role in achieving parameter control and reliable operation in industrial processes. This article proposes a more practical method for the measurement of void fraction in gas-water bubbly flow using a derived multi-eigenvalue sequence from a normalized electrical impedance tomography impedance matrix. The relations between eigenvalues and void fraction, bubble radius, number of bubbles are investigated by numerical simulations, which illustrates the superiority of using multi-eigenvalue rather than the largest eigenvalue for void fraction prediction. The nonlinear mapping between the multi-eigenvalue sequence and void fraction is established by applying the XGBoost model with a sliding window of time series. This proposed method is verified by static and dynamic experiments using a self-developed setup in our laboratory, generating stable gas-water bubbly flow with void fraction of less than 0.12. It is shown that the proposed method can predict void fraction with a relative deviation of 10%. Compared with the conventional method based on the largest eigenvalue, the proposed method efficiently improves the measurement accuracy of void fraction in gas-water bubbly flow and applicability in actual measurement of two-phase flow, which can be further extended to other flow regimes.

Distributed data-driven fault detection for industrial interconnected systems with unknown topology structure

Driven by the enhancing requirements for ensuring stable operation of complex industrial processes, this paper investigates a distributed data-driven fault detection (FD) method for industrial interconnected systems. The topology structure of the industrial interconnected system is presumed to be unknown. To this end, the input/output (I/O) data-set model for the interconnected system and the distributed input/output data-set models for the subsystems are derived first. Then, the topology structure of the interconnected system is reconstructed through data-driven realization of gap metric by utilizing the subspace Identification method. Subsequently, the distributed data-driven fault detection method with enhanced information exchange topology is developed. Finally, we validate the feasibility and efficiency of the proposed method by conducting a case study on a chemical process consisting of four continuous-stirred tank reactors (CSTRs).

Full Decoupling High-Order Dynamic Mode Decomposition for Advanced Static and Dynamic Synergetic Fault Detection and Isolation

Real industrial processes often present coupled static and dynamic characteristics, leading to significant challenges for fault detection and isolation. However, traditional dynamic modeling methods may lead to the coupling problem of statics and dynamics, which provide ambiguous process status descriptions and incorrect fault isolation results. In this work, a novel full decoupling high-order dynamic mode decomposition (FDHODMD) method is developed for fault detection and isolation of dynamic processes. Different from the existing dynamic methods, the proposed FDHODMD can separate the high-order dynamic information from static information. First, the static characteristics are separated by discarding the decomposed features corresponding to smaller singular values. Then, a high-order dynamic model is established to present the temporal relationships between variables. In this way, the effect of static characteristics on dynamics analysis can be eliminated. Accordingly, from both static and dynamic perspectives, multiple statistics are designed to comprehensively detect the anomalies and provide an explicit status identification. In addition to the static fault isolation strategy, a dynamic fault isolation strategy is designed to recognize the fault variables after detecting the deviation, which divides the complex dynamic system into several individual dynamic modes to avoid the interaction of dynamic characteristics. Thereupon, the contributions of different variables to each dynamic mode can be clearly presented to recognize the fault variables. Finally, the validity of the proposed method is illustrated through both a numerical case and a real industrial process. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In industrial processes, the static and dynamic characteristics of data are often coupled. This work presents a FDHODMD method, which can be considered the high-order version of DMD, to achieve the decoupling of dynamics and statics while extracting high-order dynamic characteristics. Then the proposed fault detection strategy designs multiple statistics for monitoring the process from both dynamic and static perspectives, which ensures the safe operation of industrial processes. After detecting the anomaly, the proposed fault isolation strategy can recognize the fault variables separately dominating the dynamic and static deviation, which may help engineers locate the fault source.

IMPLEMENTASI APLIKASI PENGENALAN SOP PENGOLAHAN LIMBAH INDUSTRI PT X BERBASIS ANDROID

The management and proper handling of industrial waste are crucial for environmental sustainability and regulatory compliance. This project presents the development and implementation of an Android-based application designed to recognize and facilitate the adherence to Standard Operating Procedures (SOPs) in the industrial waste processing operations of PT X. The application employs advanced image recognition and processing techniques to identify and validate SOPs associated with different waste processing tasks, ensuring that employees follow the prescribed guidelines accurately. By providing real-time guidance and verification, the application aims to enhance operational efficiency, reduce errors, and promote responsible waste management practices. This paper outlines the key features, technical aspects, and benefits of the application, underscoring its potential to contribute to a more sustainable and compliant industrial waste processing environment at PT X.

Design and implementation of PID based flow rate control using PLC

Industrial systems require efficient techniques to observe the stable operation of various industrial processes and to achieve optimal control. Considering the importance of industrial processes, Proportional Integral and Derivative (PID), Adaptive PID, and fuzzy logic are the most utilized control systems. Programmable Logic Controller (PLC) is a low-cost solution for the industrial processes requiring control and having the flexibility of graphical user interface. In this paper, an experimental study on flow rate control system for water flowing in a vessel is realized and implemented using Proportional (P), Proportional Derivative (PD), Proportional Integral (PI), and PID controllers with Programming Logic Controller (PLC). For optimal control, the constants for the PID controller are calculated based on Zeigler-Nichols (ZN) rules. ZN tuning rules can be used to find controller constants where the plant dynamics are not available. The experimental analysis is performed to validate the theoretical concepts. The achieved results and analysis demonstrate that the process variable, which is water inflow rate 4.92L/minute, is equal to the set point without any overshoot and remains controllable at every set point change.

Spherical-dynamic time warping – A new method for similarity-based remaining useful life prediction

Machinery prognostics and health management (PHM) plays a key role in the reliable and efficient operation of industrial processes. With the emerging big data era, data-driven prognostic methods which avoid considering complicated system models have attracted growing research interest. Among many data-driven models, similarity-based prediction methods have been popular due to their strong interpretability and relatively simple implementation process. Nevertheless, when quantifying the similarity between two trajectories, most existing similarity measures neglect the nonlinearity of the distance measurement at different degradation stages and degradation alignments with timing difference, which may not be sufficient to retrieve the most suitable trajectories for remaining useful life (RUL) prediction. To overcome these limitations, a spherical-Dynamic Time Warping (spherical-DTW) algorithm is put forward to find an optimal match between the test and training trajectories at the retrieval step. Dynamic Time Warping allows degradation alignments with timing difference through stretching or compressing the trajectories with regard to time, thereby the data in similar degradation levels can be well aligned across different units. Moreover, a newly defined nonlinear spherical distance method is introduced and incorporated into the retrieval process to account for the nonlinearity of the damage propagation process. The significance of this study is that the newly proposed spherical-DTW algorithm goes one step further to consider the nonlinearity of fault evolutions and allow degradation pattern alignments with timing difference when performing similarity-based prognostics. Two run-to-failure cases, involving a real-world industrial compressor failure case and a gas turbine engine failure dataset, are investigated to demonstrate the effectiveness and superiority of the proposed algorithm.

A classified retail pricing and production scheduling model for industrial customers

In the electricity retail market, it’s an important operation way for a retailer to provide diversified services and design appropriate retail pricing strategies that stimulate demand side responses in order to increase profit and maintain customer welfare. This paper proposes a novel retail contract design that a retailer can adopt to provide different time-of-use (TOU) prices for the various types of industrial customers based on their adjustabilities and a part of rewards from the demand response (DR) market to those who respond to the urgent electricity requirements. A bi-level optimization model of retail pricing for classified industrial customers is set up based on their adjustable capacities of production process operations. At the upper level, the retailer’s benefit is maximized considering the stochastic risks of load fluctuation and the spot market. This is achieved based on the electricity consumption strategies of classified industrial consumers, which are obtained from the lower level. Decisions are then made concerning purchases from the spot market and retail pricing strategies for industrial customers. Based on the retail prices from the upper level and industrial production processes modeling, the model at the lower level takes the maximum industrial user’s benefit as the objective to obtain the electricity consumption scheme of each type of industrial user. The model optimizes both the time division and price level of retail TOU prices, resulting in detailed electricity consumption strategies for all types of industrial customers. The results of the cases show that the proposed retail pricing can improve adjustable capacities of industrial production process operations, effectively smooth the load curve of the retailer by making full use of the peak-load-staggered among industrial customers, and boost profits of both the retailer and industrial customers. Thus, a mutually beneficial situation between the retailer and industrial customers is realized.

Support for Sustainable Planning

: Intelligent energy management is getting increasingly important for the operation of industrial processes in a globalized world and this trend is expected to continue in the future. This article shows that energy management systems will also be increasingly important for sustainable investment planning such as building or extending a site or plant. Typical investment support methods range from simple spreadsheet-based tools to accurate but resource-intensive simulation studies. However, there is often a need to compare many alternative scenarios quickly to narrow the search space early. Hence, the expectation is not to achieve the level of accuracy of comprehensive studies. Instead, the task is to be as accurate as possible using the limited data available during preliminary frontend engineering and design (preFEED). To address this challenge, a framework was developed that provides more accurate investment decisions by integrating operational aspects in the planning phase. By this, a significant increase in accuracy can be achieved, while obtaining tangible recommendations within acceptabe time. We outline the principal functionality of this framework and share our insights by means of case studies in three different industrial application domains. Finally, we describe the faced challenges and share some lessons learned.

Industrial Data-Driven Processing Framework Combining Process Knowledge for Improved Decision Making—Part 1: Framework Development

Data management systems are increasingly used in industrial processes. However, data collected as part of industrial process operations, such as sensor or measurement instruments data, contain various sources of errors that can hamper process analysis and decision making. The authors propose an operating-regime-based data processing framework for industrial process decision making. The framework was designed to increase the quality and take advantage of available process data use to make informed offline strategic business operation decisions, i.e., environmental, cost and energy analysis, optimization, fault detection, debottlenecking, etc. The approach was synthesized from best practices derived from the available framework and improved upon its predecessor by putting forward the combination of process expertise and data-driven approaches. This systematic and structured approach includes the following stages: (1) scope of the analysis, (2) signal processing, (3) steady-state operating periods detection, (4) data reconciliation and (5) operating regime detection and identification. The proposed framework is applied to the brownstock washing department of a dissolving pulp mill. Over a 5-month period, the process was found to be in steady-state 32% of the time. Twenty (20) distinct operating regimes were identified. Further processing with the help of data reconciliation techniques, principal component analysis and k-means clustering showed that the main drivers explaining the operating regimes are the pulp level in tanks, its density, and the shower wash water flow rate. Additionally, it was concluded that the top four persistently problematic sensors across the steady-state spans that would need to be verified are three flow meters (06FIC137, 06FIC152, and 06FIC433), and one consistency sensor (06NIC423). This information was relayed to process experts contacts at the plant for further investigation.

Teacher–Student Uncertainty Autoencoder for the Process-Relevant and Quality-Relevant Fault Detection in the Industrial Process

Fault detection plays an important role in process monitoring, while current fault detection methods only concentrate on process-relevant or quality-relevant faults. Therefore, in this paper, a fault detection method based on the improved teacher-student network is proposed, in which both the process-relevant and quality-relevant faults are monitored. Concretely, the student network extracts representation features and the teacher network detects faults. As the features difference between the teacher and student networks can cause performance degradation when the features of teacher network are replaced by the one from student network, representation evaluation block (REB) is proposed to evaluate and reduce the features difference. As a concrete method of REB, uncertainty modeling, is proposed to quantify the features difference and alleviate the aleatoric uncertainty, modeling features difference as a central isotropic Gaussian distribution. Then, asynchronous iterative is designed to implement teacher network and student network joint training. Accordingly, REB based on uncertainty modeling is applied in the teacher-student network named as teacher-student uncertainty autoencoder (TSUAE). Finally, a fault detection framework based on TSUAE is proposed, the effectiveness of which is verified in a wastewater treatment process. <i>Impact Statement-</i>Fault detection is a crucial technology to ensure the normal operation of industrial processes. Both quality-relevant and process-relevant faults should be concerned. The proposed fault detection framework can monitor both the quality-relevant and process-relevant subspaces based on the improved teacher-student network. To improve the accuracy of fault detection and reduce the features difference between the teacher and student networks that can cause performance degradation of the student network, the method quantifies the features difference through uncertainty modeling, reduces the features difference with representation evaluation block and asynchronous iterative training. After that, the proposed method achieves higher detection rates than traditional methods in the wastewater treatment process. Additionally, the proposed method can more efficiently detect not only the largely quality-relevant faults but also the incipient quality-relevant fault in most situations, which is significant for actual industrial processes.