Large-scale Industrial Processes Research Articles

MOLA: Enhancing Industrial Process Monitoring Using a Multi-Block Orthogonal Long Short-Term Memory Autoencoder

In this work, we introduce MOLA, a multi-block orthogonal long short-term memory autoencoder paradigm, to conduct accurate, reliable fault detection of industrial processes. To achieve this, MOLA effectively extracts dynamic orthogonal features by introducing an orthogonality-based loss function to constrain the latent space output. This helps eliminate the redundancy in the features identified, thereby improving the overall monitoring performance. On top of this, a multi-block monitoring structure is proposed, which categorizes the process variables into multiple blocks by leveraging expert process knowledge about their associations with the overall process. Each block is associated with its specific orthogonal long short-term memory autoencoder model, whose extracted dynamic orthogonal features are monitored by distance-based Hotelling’s T2 statistics and quantile-based cumulative sum (CUSUM) designed for multivariate data streams that are nonparametric and heterogeneous. Compared to having a single model accounting for all process variables, such a multi-block structure significantly improves overall process monitoring performance, especially for large-scale industrial processes. Finally, we propose an adaptive weight-based Bayesian fusion (W-BF) framework to aggregate all block-wise monitoring statistics into a global statistic that we monitor for faults. Fault detection speed and accuracy are improved by assigning and adjusting weights to blocks based on the sequential order in which alarms are raised. We demonstrate the efficiency and effectiveness of our MOLA framework by applying it to the Tennessee Eastman process and comparing the performance with various benchmark methods.

Enrichment optimization of Milk Fat Globule Membrane extracts from buttermilk using Response Surface Methodology

This study investigates the effect of rennet-induced coagulation on the composition of Milk Fat Globule Membrane (MFGM) extracts from buttermilk using Response Surface Methodology. Optimizing coagulation conditions, such as rennet concentration, incubation time and temperature, significantly affects the protein and lipid fractions of the resulting MFGM extracts. Certain treatments yielded promising results in terms of MFGM concentration and casein removal. Principal Component Analysis showed different compositional patterns in the samples, shedding light on the intricate interaction of factors throughout the coagulation process. Buttermilk wheys (BMWs) characterized by high levels of both lipid and protein MFGM components, are of interest for practical applications. Although the extract with the highest MFGM protein (43.60 g/100 mL) and a medium content of polar lipids (12.98 g/100 g of fat) was found in sample C35min40°C, other BMW, such as those from the treatments C60min30°C, E35min30°C or C35min20°C, were also sufficiently enriched in MFGM associated compounds. In conclusion, the study highlights the importance of tailored coagulation conditions for optimizing the nutritional and functional properties of MFGM extracts, offering valuable insights for further nutritional research and potential applications in large-scale industrial processes.

Data-Driven Fault Classification in Large-Scale Industrial Processes Using Reduced Number of Process Variables

Data-Driven Fault Classification in Large-Scale Industrial Processes Using Reduced Number of Process Variables

Development of cost-effective cellulose-based bilayer hybrid composite membranes for CO2 separation in biogas purification

CO2 is the main pollutant in biogas, reducing its calorific value. Among various technological methods to eliminate carbon dioxide from biogas, membrane separation technology stands out for large-scale industrial biogas purification due to its advantages. The selection of membrane material and preparation process are key factors in membrane separation technology. In this study, a premixing process was initially used to blend different masses of packed molecular sieves TS-1 (or ZSM-5) and cellulose derivatives (ethyl cellulose, cellulose acetate) separately. These mixtures were then coated onto porous PVDF substrates using a coating process to create various bilayer hybrid composite membranes. Among these, PVDF/EC-ZSM-5 (containing 15 % ZSM-5) bilayer hybrid composite membrane is the most fitting. For CO2/CH4 gas mixtures, the gas selectivity of this membrane surpassed Robeson's 1991 standard line (the CO2 permeability was 597.48 Barrer, and the CO2/CH4 selectivity was 9.14). Overall, this composite membrane, made for the first time from PVDF, ZSM-5, and EC, is expected to be a promising CO2 selective separation membrane material for biogas purification in large-scale industrial processes due to its simple production process.

Integrating supervised and unsupervised learning approaches to unveil critical process inputs

This study introduces a machine learning framework tailored to large-scale industrial processes characterized by a plethora of numerical and categorical inputs. The framework aims to (i) discern critical parameters that influence the output and (ii) generate accurate out-of-sample qualitative and quantitative predictions of production outcomes. Specifically, we address the pivotal question of the significance of each input in shaping the process outcome, using an industrial Chemical Vapor Deposition (CVD) process as an example. The initial objective involves merging subject matter expertise and clustering techniques exclusively on the process output, here, coating thickness measurements at various positions in the reactor. This approach identifies groups of production runs that share similar qualitative characteristics, such as film mean thickness and standard deviation. In particular, the differences of the outcomes represented by the different clusters can be attributed to differences in specific inputs, indicating that these inputs are potentially critical to the production outcome. Shapley value analysis corroborates the formed hypotheses. Leveraging this insight, we subsequently implement supervised classification and regression methods using the identified critical process inputs. The proposed methodology proves to be valuable in scenarios with a multitude of inputs and insufficient data for the direct application of deep learning techniques, providing meaningful insights into the underlying processes.

Mössbauer spectroscopy in studying industrial catalysts and processes in recycling

Advantageous properties of Mössbauer spectroscopy allow the method to be used for the examination of industrial catalysts as well. The method is confined for catalysts containing Mössbauer nuclei (57Fe and 119Sn in most practical cases), on the other side the obtained information is rather unique. Some further limitations emerge in studies for industrial catalysts with this technique in comparison to common catalyst studies, namely the extreme operating conditions, i.e. the elevated temperature and pressure. Thus the in situ conditions for studies cannot be completed easily. To circumvent this difficulty the states of catalysts prior and after the usage can be compared (ex situ conditions). Further, since real catalysts are used for long time period in the industry, their studies can be replaced by accelerated studies performed under ‘conditions relevant to real processes’. Beside catalysis, recycling and reuse of former industrial waste is also an important field for application of the method. A personal selection from reports from recent years on application of the method in representative large scale industrial processes is presented in the overview.

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.

Fault detection of industrial processes using attention-based gated recurrent unit autoencoder with skip connection

Abstract With the continuous evolution of modern industrial technology, industrial production has grown progressively complex, necessitating the use of various sensors to measure multiple process variables. However, intricate temporal dependencies and nonlinear relationships between data presented by multivariate sequences pose significant challenges to process fault detection. In response to these challenges, this paper proposes an attention-based gated recurrent unit autoencoder with skip connection (SAGRU-AE) model for monitoring large-scale, nonlinear, and multivariate industrial process faults. SAGRU-AE combines gated recurrent units, multi-head self-attention, and autoencoder to extract features from multivariable time series data efficiently. Concurrently, feature reuse is achieved through the skip connection structure, which improves the accuracy of data reconstruction. Based on the implementation of process data feature extraction and input reconstruction in SAGRU-AE, two statistics have been developed, namely the H2 statistic and square prediction error (SPE) statistic, for fault detection tasks. Ultimately, the feasibility and effectiveness of the proposed algorithms are validated through experimentation on the TE process.

Bridging papermaking and hydrogel production: Nanoparticle-loaded cellulosic hollow fibers with pitted walls as skeleton materials for multifunctional electromagnetic hydrogels

Electromagnetic hydrogels have attracted significant attention due to their vast potential in soft robotics, biomedical engineering, and energy harvesting. To facilitate future commercialization via large-scale industrial processes, we present a facile concept that utilizes the specialized knowledge of papermaking to fabricate hydrogels with multifunctional electromagnetic properties. The principles of papermaking wet end chemistry, which involves the handling of interactions among cellulosic fibers, fines, polymeric additives, and other components in aqueous systems, serves as a key foundation for this concept. Notably, based on these principles, the versatile use of chemical additives in combination with cellulosic materials enables the tailored design of various products. Our methodology exploits the unique hierarchically pitted and hollow tube-like structures of papermaking grade cellulosic fibers with discernible pits, enabling the incorporation of magnetite nanoparticles through lumen loading. By combining microscale softwood-derived cellulosic fibers with additives, we achieve dynamic covalent interactions that transform the cellulosic fiber slurry into an impressive hydrogel. The cellulosic fibers act as a skeleton, providing structural support within the hydrogel framework and facilitating the dispersion of nanoparticles. In accordance with our concept, the typical hydrogel exhibits combined attributes, including electrical conductivity, self-healing properties, pH responsiveness, and dynamic rheologic behavior. Our approach not only yields hydrogels with interesting properties but also aligns with the forefront of advanced cellulosic material applications. These materials hold the promise in remote strain sensing devices, electromagnetic navigation systems, contactless toys, and flexible electronic devices. The concept and findings of the current work may shed light on materials innovation based on traditional pulp and paper processes. Furthermore, the facile processes involved in hydrogel formation can serve as valuable tools for chemistry and materials education, providing easy demonstrations of principles for university students at different levels.

Tailoring the Structure-Property Relationship of Ring-Opened Metathesis Copolymers for CO2-Selective Membranes.

In this work, we explore the use of ring-opening metathesis polymerization (ROMP) facilitated by a second-generation Grubbs catalyst (G2) for the development of advanced polymer membranes aimed at CO2 separation. By employing a novel copolymer blend incorporating 4,4'-oxidianiline (ODA), 1,6-hexanediamine (HDA), 1-adamantylamine (AA), and 3,6,9-trioxaundecylamine (TA), along with a CO2-selective poly(ethylene glycol)/poly(propylene glycol) copolymer (Jeffamine2003) and polydimethylsiloxane (PDMS) units, we have synthesized membranes under ambient conditions with exceptional CO2 separation capabilities. The strategic inclusion of PDMS, up to a 20% composition within the PEG/PPG matrix, has resulted in copolymer membranes that not only surpass the 2008 upper limit for CO2/N2 separation but also meet the commercial targets for CO2/H2 separation. Comprehensive analysis reveals that these membranes adhere to the mixing rule and exhibit percolation behavior across the entire range of compositions (0-100%), maintaining robust antiplasticization performance even under pressures up to 20 atm. Our findings underscore the potential of ROMP in creating precisely engineered membranes for efficient CO2 separation, paving the way for their application in large-scale environmental and industrial processes.

A novel process monitoring framework combined temporal feedback autoencoder and multilevel correlation analysis for large-scale industrial processes

Modeling and monitoring generally face numerous challenges such as complex characteristics of multi-unit, temporal correlations and strong interaction among subblocks in large-scale industrial processes. To handle those challenges, a novel process monitoring framework combined temporal feedback autoencoder and multilevel correlation analysis is proposed in this paper. Firstly, large-scale industrial processes are decomposed into multiple subblocks in spatial and temporal order based on process knowledge. Secondly, an improved autoencoder with temporal feedback mechanism is constructed as local monitoring model to capture the important latent variables of each subblock. Then, considering the sequential transmission and correlations among subblocks in series, a multilevel correlation analysis method is employed to efficiently extract the unique features of each subblock and the joint features of the whole process. Finally, the irregular contribution indices of the unique features and the joint features are designed for hierarchical process monitoring. The superiority of the proposed framework can be verified by Tennessee Eastman process and a real hot strip mill process.

When grey model meets deep learning: A new hazard classification model

The classification of hazard is critical in industrial informatics, as it enhances early safety alerts, supports decision-making, and facilitates policy assessment. However, previous studies have generally neglected the temporal attributes of hazards, thereby constraining the effectiveness of models. This paper introduces a new model for hazard classification termed DLGM. DLGM represents a deep learning framework, with the structural parameters of grey models to encapsulate the hazard temporal attributes. To better accommodate the fluctuations of hazard series, a new grey model termed FSGM(1,1) equipped with Fourier series is proposed. Moreover, DLGM leverages a novel hierarchical feature fusion neural network (HFFNN) to optimize feature processing. Extensive experiments across three hazard themes involving 18 large-scale industrial processes have demonstrated the competitiveness of DLGM (e.g., it surpasses benchmark models such as Random Forest and BERT by approximately 2% in both accuracy and F1), the suitability of FSGM(1,1) (e.g., its mean absolute percentage error is less than 10% and mean squared error is below 0.02) and the effectiveness of HFFNN (e.g., it enhances the accuracy and F1 of DLGM by about 1%).

Distributed incipient fault detection with causality-based multi-perspective subblock partitioning for large-scale nonlinear processes

Detecting incipient faults is a crucial task in complex large-scale nonlinear industrial processes. However, conventional distributed frameworks encounter challenges in identifying appropriate subblock partitioning methods. To address this problem, we propose a novel distributed incipient fault detection method, called CMPSP-MKPCA, based on causal relationship multi-perspective subblock partitioning (CMPSP) and modified kernel principal component analysis (MKPCA), facilitating incipient fault detection in large-scale nonlinear processes. Firstly, considering the property of incipient faults to undergo limited propagation among process flow connections, we obtain fine-grainehe direction of potential incipient fault information propagation. Secondly, the design of multi-perspective detection blocks provides real-time selection of sub-block delineation results to adapt to the different needs of incipient faults during latency and transition periods. Thirdly, local detection is realized by combining the Modified KPCA method and Kullback-Leibler divergence (KLD), while the global detection statistics is obtained by Bayesian inference fusion. Simulation experiments based on the continuous stirred tank reactor (CSTR) system and the Tennessee Eastman (TE) system verify the effectiveness and superiority of the method.

Multi-node knowledge graph assisted distributed fault detection for large-scale industrial processes based on graph attention network and bidirectional LSTMs

Modern industrial processes are characterized by extensive, multiple operation units, and strong coupled correlation of subsystems. Fault detection of large-scale processes is still a challenging problem, especially for tandem plant-wide processes in multiple fields such as water treatment process. In this paper, a novel distributed graph attention network-bidirectional long short-term memory (D-GATBLSTM) fault detection model is proposed for large-scale industrial processes. Firstly, a multi-node knowledge graph (MNKG) is constructed using a joint data and knowledge driven strategy. Secondly, for large-scale industrial process, a global feature extractor of graph attention networks (GATs) is constructed, on the basis of which, sub-blocks are decomposed based on MNKG. Then, local feature extractors of bidirectional long short-term memory (Bi-LSTM) for each sub-block are constructed, in which correlations among multiple sub-blocks are considered. Finally, a multi-subblock fusion collaborative prediction model is constructed and the comprehensive fault detection results are given by the grid search method. The effectiveness of our D-GATBLSTM is exemplified in a secure water treatment process case, where it outperforms baseline models compared, with 27% improvement in precision, 15% increase in recall, and overall F-score enhancement of 0.22.

Insertion of Molecular Oxygen into a Gold(III)-Hydride Bond.

The use of molecular oxygen as an oxidant in chemical synthesis has significant environmental and economic benefits, and it is widely used as such in large-scale industrial processes. However, its adoption in highly selective homogeneous catalytic transformations, particularly to produce oxygenated organics, has been hindered by our limited understanding of the mechanisms by which O2 reacts with transition metals. Of particular relevance are the mechanisms of the reactions of oxygen with late transition metal hydrides as these metal centers are better poised to release oxygenated products. Homogeneous catalysis with gold complexes has markedly increased, and herein we report the synthesis and full characterization of a rare AuIII-H, supported by a diphosphine pincer ligand (tBuPCP = 2,6-bis(di-tert-butylphosphinomethyl)benzene). [(tBuPCP)AuIII-H]+ was found to cleanly react with molecular oxygen to yield a stable AuIII-OOH complex that was also fully characterized. Extensive kinetic studies on the reaction via variable temperature NMR spectroscopy have been completed, and the results are consistent with an autoaccelerating radical chain mechanism. The observed kinetic behavior exhibits similarities to that of previously reported PdII-H and PtIV-H reactions with O2 but is not fully consistent with any known O2 insertion mechanism. As such, this study contributes to the nascent fundamental understanding of the mechanisms of aerobic oxidation of late metal hydrides.

Directed evolution of highly sensitive and stringent choline-induced gene expression controllers.

Gene expression controllers are useful tools for microbial production of recombinant proteins and valued bio-based chemicals. Despite its usefulness, they have rarely been applied to the practical industrial bioprocess, due to the lack of systems that meets the three requirements: low cost, safety, and tight control, to the inducer molecules. Previously, we have developed the high-spec gene induction system controlled by safe and cheap inducer choline. However, the system requires relatively high concentration (~100 mM) of choline to fully induce the gene under control. In this work, we attempted to drastically improve the sensitivity of this induction system to further reduce the induction costs. To this end, we devised a simple circuit which couples gene induction system with positive-feedback loop (P-loop) of choline importer protein BetT. After the tuning of translation level of BetT (strength of the P-loop) and deletion of endogenous betI (noise sources), highly active yet stringent control of gene expression was achieved using about 100 times less amount of inducer molecules. The choline induction system developed in this study has the lowest basal expression, the lowest choline needed to be activated, and the highest amplitude of induction as the highest available promoter such as those known as PT5 system. With this system, one can tightly control the expression level of genes of interest with negligible cost for inducer molecule, which has been the bottleneck for the application to the large-scale industrial processes.

An Overview of Heterogeneous Catalysts Based on Hypercrosslinked Polystyrene for the Synthesis and Transformation of Platform Chemicals Derived from Biomass.

Platform chemicals, also known as chemical building blocks, are substances that serve as starting materials for the synthesis of various value-added products, which find a wide range of applications. These chemicals are the key ingredients for many fine and specialty chemicals. Most of the transformations of platform chemicals are catalytic processes, which should meet the requirements of sustainable chemistry: to be not toxic for humans, to be safe for the environment, and to allow multiple reuses of catalytic materials. This paper presents an overview of a new class of heterogeneous catalysts based on nanoparticles of catalytically active metals stabilized by a polymer matrix of hypercrosslinked polystyrene (HPS). This polymeric support is characterized by hierarchical porosity (including meso- and macropores along with micropores), which is important both for the formation of metal nanoparticles and for efficient mass transfer of reactants. The influence of key parameters such as the morphology of nanoparticles (bimetallic versus monometallic) and the presence of functional groups in the polymer matrix on the catalytic properties is considered. Emphasis is placed on the use of this class of heterogeneous catalysts for the conversion of plant polysaccharides into polyols (sorbitol, mannitol, and glycols), hydrogenation of levulinic acid, furfural, oxidation of disaccharides, and some other reactions that might be useful for large-scale industrial processes that aim to be sustainable. Some challenges related to the use of HPS-based catalysts are addressed and multiple perspectives are discussed.

Material Challenges and Hydrogen Embrittlement Assessment for Hydrogen Utilisation in Industrial Scale

Hydrogen has been studied extensively as a potential enabler of the energy transition from fossil fuels to renewable sources. It promises a feasible decarbonisation route because it can act as an energy carrier, a heat source, or a chemical reactant in industrial processes. Hydrogen can be produced via renewable energy sources, such as solar, hydro, or geothermic routes, and is a more stable energy carrier than intermittent renewable sources. If hydrogen can be stored efficiently, it could play a crucial role in decarbonising industries. For hydrogen to be successfully implemented in industrial systems, its impact on infrastructure needs to be understood, quantified, and controlled. If hydrogen technology is to be economically feasible, we need to investigate and understand the retrofitting of current industrial infrastructure. Currently, there is a lack of comprehensive knowledge regarding alloys and components performance in long-term hydrogen-containing environments at industrial conditions associated with high-temperature hydrogen processing/production. This review summarises insights into the gaps in hydrogen embrittlement (HE) research that apply to high-temperature, high-pressure systems in industrial processes and applications. It illustrates why it is still important to develop characterisation techniques and methods for hydrogen interaction with metals and surfaces under these conditions. The review also describes the implications of using hydrogen in large-scale industrial processes.

Automatic decomposition of large-scale industrial processes for distributed MPC on the Shell–Yokogawa Platform for Advanced Control and Estimation (PACE)

The kernel of industrial advanced process control (APC) lies in the formulation and solution of model predictive control (MPC) problems, which specify the controller moves according to the solution of an optimal control problem at each sampling time. A significant challenge is the online computation for large-scale industrial systems. As the state-of-the-art APC technology, the Shell–Yokogawa Platform for Advanced Control and Estimation (PACE) has adopted a systematic framework of handling dynamic optimization of large-scale systems, where an automatic decomposition procedure generates subsystems for distributed MPC. The decomposition is implemented on network representations of the MPC models that capture interactions among process variables, with community detection used to maximize the statistical significance of the subnetworks with preferred internal interconnections. This paper introduces the fundamentals of such a decomposition approach and this functionality in PACE, followed by a case study on a crude distillation process to showcase its industrial application.

Multi-scenario and multi-stage robust NMPC with state estimation application on the Tennessee-Eastman process

This study presents the implementation of two discrete robust approaches to Non-linear Model Predictive Control (NMPC), multi-scenario NMPC (MSc-NMPC) and multi-stage NMPC (MS-NMPC), to the benchmark Tennessee-Eastman (TE) challenge, with Extended Kalman Filter (EKF) and Moving Horizon Estimation (MHE) as state estimators. The robust NMPC formulation results in closed-loop responses that prevent constraint violation and closely track the process set-point under parameter uncertainty, even in scenarios where traditional NMPC results in an unstable response for this process. Additionally, unconstrained state estimators such as EKF are unsuitable because the parameter uncertainty may cause estimates to fall outside the feasible region of the process, which ultimately destabilizes the process. MHE was able to overcome this challenge because it considers process constraints in its formulation. The additional computational time required to solve the robust NMPC formulations and MHE does not cause significant delays for the sampling time considered, demonstrating their applicability to challenging large-scale industrial chemical processes.