Operation Of Industrial Processes Research Articles

Surrogate‐based optimization of a periodic rescheduling algorithm

AbstractPeriodic rescheduling is an iterative method for real‐time decision‐making on industrial process operations. The design of such methods involves high‐level when‐to‐schedule and how‐to‐schedule decisions, the optimal choices of which depend on the operating environment. The evaluation of the choices typically requires computationally costly simulation of the process, which—if not sufficiently efficient—may result in a failure to deploy the system in practice. We propose the continuous control parameter choices, such as the re‐optimization frequency and horizon length, to be determined using surrogate‐based optimization. We demonstrate the method on real‐time rebalancing of a bike sharing system. Our results on three test cases indicate that the method is useful in reducing the computational cost of optimizing an online algorithm in comparison to the full factorial sampling.

Flow characteristic of steam jet condensed into a water pipe flow - a numerical study

• Numerical simulation of steam jet condensation in subcooled water flow is conducted. • Thermal hydraulic parameters of cone and cylinder-cone steam plumes are analyzed. • The total pressure in the centerline reaches maximum at steam volume fraction 0.01. • Heat transfer coefficients based on area and volume of two-phase region are obtained. Gas-liquid two-phase jet condensation has a property of high heat and mass transfer efficiency, thereby its theory and method becoming an urgent requirement for optimization design and safe operation of various industrial processes. Based on the experimental study of steam jet condensation in subcooled water flow in a vertical pipe, we have developed a numerical simulation method of steam-water two-phase jet condensation coupled with thermal phase change model and two-fluid model. Meanwhile, we emphasize on the shape of the steam plume, the distribution of pressure and velocity, and the heat and mass transfer of the two-phase interface. As the steam mass flux increases from subsonic to sonic, the steam plume becomes longer, and its shape transforms from cone to cylinder-cone gradually. The steam volume fraction is close to 0.01 at the end of the two-phase region, where the total pressure in the centerline approaches the total pressure of the water and reaches the maximum. The mass transfer process of steam condensed into liquid mainly occurs in the gas–liquid two-phase region, accompanied by drastic parameter change in pressure and velocity. Under the working conditions of this paper, the heat transfer coefficients based on the interfacial area of the steam plume and the volume of the two-phase region range from 1.71 MW/m 2 K to 3.51 MW/m 2 K, and from 2398.80 MW/m 3 K to 5040.07 MW/m 3 K, respectively.

Stable soft sensor modeling based on causality analysis

Data-driven soft sensors, aiming to estimate and predict hard-to-measure quality variables using easy-to-measure process variables, have now become the key foundation for monitoring the stable and safe operation of industrial processes. However, traditional machine-learning methods usually make an assumption that training data and test data share the same probability distribution or the probability distribution of test data is known, which is impractical in the fact that test data come from multi-unknown operating modes. Based on causality analysis and stable learning, soft sensors for stable prediction, namely stable soft sensors, are proposed in this paper. To address this problem, three stable soft sensor frameworks based on causal variables, unsupervised causal features, and supervised causal features are designed. By introducing causality in soft sensor modeling, the interpretability is enhanced and the prediction results in different operating modes get stable. The effectiveness of the proposed method is shown through case studies in the benchmark Tennessee Eastman process.

Optimizing Resource Conservation Networks: Overcoming Barriers with Process Systems Engineering

Climate change continues to be the major environmental challenge of this century. Some of its effects may include more severe weather conditions, increased incidence of water scarcity, and the potential of pandemic outbreaks, to name a few. Scientific research has thus looked into various strategies to address this problem. These include reducing the generation of greenhouse gas emissions (GHGs) through the implementation of low carbon technologies and transitioning towards a more circular economy through improved resource efficiency. The challenge is in identifying which among the available options would yield the best results. The field of process systems engineering (PSE) which is a computational branch of chemical engineering, provides rigorous methods for analyzing systems which can be used to support decision making. These methodologies have been implemented at various scales to identify strategies which can improve the environmental performance of industrial systems. It has been utilized for optimizing operations of industrial processes, managing resources in eco-industrial parks, greening the product supply chain, and even analyzing strategies for reducing national carbon emissions. This work provides an overview of how PSE tools can be utilized for the design of more sustainable systems and showcases how it can be used at various levels of implementation.

Sliding Mode Observer Design for decentralized multi-phase flow estimation

Robust flow measurement in multi-phase systems has extensive applications in understanding, design, and operation of complex environmental, energy and industrial processes. The nonlinearity and spatiotemporal variability of the interactions between different flow phases makes the multi-phase flow measurement a challenging task. Two Sliding Mode Observer (SMO) schemes are proposed in this study for the state estimation of a decentralized multi-phase flow measurement system. The developed observers are shown to be theoretically valid and numerically applicable for a real-life case study data. The multi-phase flow system considered in this paper can be described as two interconnected sub-systems including fluid and gas sub-systems, and two scenarios are considered in the design of the observers. The first scenario considers the interconnections as bounded disturbance (SMOD), while the second scenario considers the interconnections as an uncertainty (SMOU). Hence, the Sliding Mode Observers are adopted to mitigate the effects of disturbance in the system and uncertainties in the parameters. Numerical simulations are conducted using MATLAB and dynamic HYSYS simultaneously, using the data obtained from field-based multi-phase flow measurements. The results demonstrate the appropriateness and robustness of the proposed Sliding Mode Observer (SMO) for estimation of the multi-phase fluid specifications including the density, velocity, and the volume phases fraction in each subsystem. The analysis of the results highlights that the proposed model is computationally efficient with fast transient response, accurate tracking capability of the real process data, and very low steady-state error. This study shows that choosing an appropriate Lyapunov-Krasovsky function results in the asymptotic stability of the decentralized system and improves the performance of the proposed observers. Uncertainty analysis is conducted on the velocity estimation results obtained from the Sliding Observers. Overall, SMOU method shown better performance with RMSE of 0.24%, while RMSE of 0.46% was achieved for the SMOD. Comparison of the numerical results with the field-based flow measurement, as a benchmark, shows that although uncertainty in SMOU is approximately half of the uncertainty in SMOD, state estimation for both schemes was achieved in a finite time with high order of precision. It was shown that both observers developed in this study are well capable of estimating the multi-phase flow variables and states.

Flexible robotic strategy for the assembly of ring-shaped elastic objects

This paper addresses the challenge of the robotized insertion of ring-shaped elastic objects into parts’ outer grooves, which is a key operation in several industrial processes. We propose a novel flexible strategy to insert ring-shaped elastic objects of different sizes and stiffnesses, exploiting a generic robot and a simple - yet effective - fixture. Despite the overall complexity of the procedure is entirely poured on the mentioned fixture, its simplicity of realization and implementation allows to equip the robot with a generic gripper, without the use of any additional sensor. The effectiveness of the proposed strategy is assessed via experimental tests, proving the efficiency and flexibility in the insertion of different ring-shaped elastic objects.

Incipient Sensor Fault Detection by Directly Monitoring Sliding Window Based Singular Values*

Timely and accurate detection of incipient faults is critical to guarantee the normal operation of industrial processes. Nowadays, complex systems are usually equipped with a large number of sensors, which may be vulnerable to faults due to harsh environments. Statistical process monitoring is commonly used for fault detection purpose. Nevertheless, traditional fault detection methods are not sensitive enough to incipient faults, leading to the occurrence of many missed alarms. In this paper, the incipient fault detection task is achieved by monitoring the changes of sample singular values within a sliding window. Two incipient sensor fault types are considered, i.e. the sensor constant bias fault and sensor precision degradation fault. In addition, the rationale behind this strategy is also theoretically analyzed. Finally, a numerical example and the continuous stirred tank reactor process demonstrate the effectiveness of the proposed method.

Don’t get stung, cover your ICS in honey: How do honeypots fit within industrial control system security

The advent of Industry 4.0 and smart manufacturing has led to an increased convergence of traditional manufacturing and production technologies with IP communications. Legacy Industrial Control System (ICS) devices, now interconnected via public networks, are exposed to a wide range of previously unconsidered threats, which must be considered to ensure the continued safe operation of industrial processes. This paper surveys the ICS honeypot deployments in the literature to date, provides an overview of ICS focused threat vectors, and studies how honeypots can be integrated within an organisations defensive strategy. We discuss relevant legislation, such as the UK Cyber Assessment Framework, the US NIST Framework for Improving Critical Infrastructure Cybersecurity, and associated industry-based standards and guidelines supporting operator compliance. This is used to frame a discussion on our survey of existing ICS honeypot implementations, and the role of honeypots in supporting regulatory objectives. We observe that many low-interaction honeypots are limited in their use. This is largely due to the increased knowledge attackers have on how real-world ICS devices are configured and operate vs the configurability of simulated honeypot systems. Furthermore, we find that environments with increased interaction provide more extensive capabilities and value, due to their inherent obfuscation delivered through the use of real-world systems. Based on these insights, we propose a novel framework towards the classification and implementation of ICS honeypots.

A Survey on Various Attacks and Countermeasures in Wireless Sensor Networks

The Wireless Sensor Network (WSN) is an evolving technological base comprising spatially dispersed, autonomous tiny sensing devices called nodes. Today, wireless sensor networks are not exclusively limited to military uses such as frontline surveillance. However, they are still used in various industrial and civilian applications, including industrial process tracking and operation, robotic health monitoring, vibration and flame monitoring, healthcare systems, home automation, and traffic management. Nodes, such as temperature, hermetic, vibration, pressure, leisure seizure, or pollutant parameters, are deployed independently to cooperatively control mammal or environmental conditions. The compulsion of security issues occurs in the Wireless Sensor Network to be accepted with complete zip honesty, confidentiality, authentication during contact. In this paper, we analyze the security criteria, internal and external security threats and attacks that can be done, and the mechanism used in the Wireless Sensor Network to address such security issues.

Biowaste-based biodegradable flocculants for clean and sustainable tailings management in industrial mining and mineral processing

Massive fluid tailings discharged from industrial mining and mineral processing operations pose numerous economic and environmental challenges. To remove water and condense tailings into sediment, we developed a novel bioflocculant from specified risk materials, a particular type of waste biomass from the animal rendering industry. By properly controlling key synthesis conditions, the bioflocculant successfully met industrial requirements for fluid tailings treatment, including enhanced short-term tailings settling and improved ultimate water removal extent. Settling tests using real industrial fluid tailings showed that the synthesized bioflocculant enabled an initial settling rate 4 magnitudes higher than the control case and achieved 93.2% the dewatering efficiency of a commercial flocculant (hydrolyzed polyacrylamide; HPAM). In contrast to HPAM, our bioflocculant can achieve tailings settling without gypsum, a common processing aid that can hinder downstream industrial operations. More importantly, growth tests using two common microorganisms (Saccharomyces cerevisiae and Escherichia coli) suggested that the bioflocculant can be easily degraded, indicating reduced negative environmental impacts after its use. Taken together, this study demonstrates the great potential of implementing a biodegradable bioflocculant from proteinaceous wastes for clean and sustainable management of mining and mineral processing fluid tailings.

One-dimensional residual convolutional auto-encoder for fault detection in complex industrial processes

Fault detection and diagnosis have always been the key techniques for safe and reliable operation of industrial processes. However, the high dimension and noise of process variables have brought great challenges to the fault detection model. In recent years, due to the powerful feature extraction ability, deep learning has been widely applied in process fault detection and diagnosis. However, these deep neural networks (DNNs) often need a large amount of label data for supervised training, or show poor performance in learning features under unsupervised-learning condition. This paper proposes a new DNN model, a one-dimension residual convolutional auto-encoder (1DRCAE), where unsupervised learning is used to extract representative features from complex industrial processes. 1DRCAE effectively integrates the one-dimensional convolutional kernel with an auto-encoder and is embedded residual learning block for effective feature extraction from one-dimensional data. The two statistics and squared prediction error are generated in the feature space and residual space of 1DRCAE, respectively. Finally, the feasibility and superiority of 1DRCAE are verified on a simulation process, Tennessee Eastman process, Fed-batch fermentation penicillin process, and a real-life case. The convolutional auto-encoder technique provides a new way for feature learning and fault detection on complex industrial processes.

Investigating occupational and operational industrial safety data through Business Intelligence and Machine Learning

Learning from previous events represents a crucial element to improve the design and operations of industrial processes, especially considering the many variables characterizing the functioning of a plant. This learning process aims to reduce the frequency of incidents and/or mitigate their severity, which are both continuous and open challenges.This paper is grounded on a large incident repository, i.e., the Major Hazard Incident Data Service (MHIDAS) database, which was developed in 1986 by the Health and Safety Executive (HSE) to provide a reliable source of data on major hazard incidents involving hazardous materials. The database includes more than 9000 reports collected over five decades (1950s–1990s). This paper aims to provide a novel understanding of the industrial incidents reported in MHIDAS and unveil possible ways of exploring occupational/operational incidents through descriptive and quantitative analyses. Consequently, this paper proposes the implementation of Business Intelligence (BI) tools to facilitate dynamic data visualization and Machine Learning (ML) algorithms for the extraction of knowledge from different data entries. Therefore, after engineering the MHIDAS data model, a set of BI dashboards was designed and complemented with a ML-driven categorization of incidents through representative key variables for occupational/operational incidents.The manuscript describes the process necessary to create a BI model for safety data management in an industrial context, and its integration with ML solutions that may support an in-depth multi-variate investigation of reported data. The investigation provides evidence on the importance of a precise reporting of safety events, thus unveiling the potential for lessons learned in the process industry.

Predicting the flowability of alumina powder during batch grinding through the establishment of a grinding kinetic model

Grinding is one of the main unit operation in industrial processes handling powders. The particle size reduction that takes place during grinding tests, usually results in a significant change in the flow behavior of the ground powder. Up to now, a model predicting the evolution of powder flowability with grinding time, according to the operating conditions is still missing. In this paper, a methodology combining a grinding kinetic model and a flowability model involving the population-dependent granular Bond number is developed. The methodology has been applied to an alumina powder, ground in a batch ball mill. The flow function coefficient of the ground samples is measured after various grinding times in a powder shear tester. The comparison between model predictions and experimental data shows that this method allows an accurate prediction of the powder flow behavior over the first sixteen minutes of grinding.

Cost-benefit analysis of calibration model maintenance strategies for process monitoring

The long-term prediction performance of spectroscopic calibration models is a critical factor to monitor or control many production processes. Over time, new variations may emerge that deteriorate prediction performance. Therefore, models have to be maintained to retain or improve their prediction performance through time, requiring considerable resources and data. Maintenance should improve relevant predictions but also needs to be resource and cost efficient. Current approaches do not consider these trade-offs. We propose a new method to quantify the effectiveness and cost of model maintenance strategies based on historical data. Model performance over time for past, imminent and future samples is evaluated as these may react differently to maintenance. The model performance and required updating resources are translated into relative cost and benefit to compare strategies and determine optimal maintenance parameters. We used this method to evaluate a maintenance strategy that combines adding incoming samples to the calibration data with re-optimization of spectral preprocessing and modelling parameters. Continuously adding samples to the calibration data is shown to improve prediction performance and leads to more robust and generic models for emerging variations in all investigated data streams. Selectively adding incoming sample variations showed a reduced prediction performance but saves considerably in resources. Comparing model performance on the different sampling windows can also be used to determine an optimal updating frequency. This novel strategy to evaluate the expected performance and determine an optimal maintenance strategy is generally applicable and should lead to robust and consistently high prospective and/or retrospective model performance through time, which can be crucial for optimal operation and fault detection in industrial processes.

Local coordinates and global structure preservation for fault detection and diagnosis

Fault detection is crucial for the safe and reliable operation of industrial processes. Manifold learning methods have been widely used by the fault detection community. However, traditional manifold learning-based fault detection methods extract either the global structure information of the given data or the local structure information hidden in the data without capturing both global and local representations, which probably deteriorates the detection capabilities of these methods. This paper proposes a new feature extraction method named local coordinates and global structure preservation (LCGSP) for fault detection, where the local tangent space information and global geometric structure of data are captured simultaneously. A distance ratio is proposed and then integrated into principal component analysis to approximate the local tangent space, and this technique can improve the accuracy of the approximated local tangent space even if the data are sparse or non-uniformly distributed. In this context, the proposed method can better preserve the local manifold structure through the use of an accurate local tangent space. Furthermore, the geodesic distances between non-neighbors are introduced to preserve the global intrinsic relationships among data points. In this way, more faithful representations of the original data can be captured, thus enhancing the detection performance of the proposed method. In addition, a Bayesian fault diagnosis strategy is developed to make decisions about fault statuses. Two case studies demonstrate the superiority of the proposed LCGSP-based fault detection and diagnosis approach.

Deployment and Evaluation of an Industry 4.0 Use Case over 5G

The arrival of 5G paves the way for the deployment of the so-called Industry 4.0, which is a new paradigm devoted to the digital transformation of manufacturing and factory production. Because of the resources required to perform this transformation, the importance of field trials and experimentation cannot be overstated, both to support the design of novel methodologies and to validate these designs. In this article, we leverage the 5G EVE end-to-end open platform to design and validate a novel operation approach for automated guided vehicles (AGVs). This use case consists of the placement of the intelligence that controls the AGV in a remote entity. This movement could improve and simplify the operation of industrial processes. The customizability of the 5G validation platform proves fundamental to evaluate the solution under different deployment architectures and to assess its performance under hazardous radio conditions. Our results demonstrate the ability of 5G to handle latency-constrained use cases with superior performance compared to the current state-of-the-art mobile technology.

Survey for Cadmium Concentration in Liver, Kidney and Muscle of Slaughtered Sheep in Butcher Shops of Fallujah City, Iraq

Cadmium (Cd) is regarded as one of major the presence of hazardous substances in the environment, with a broad variety toxicity to organs and an extended half-life of removal. Cd exposure for farmed ruminants can come from industrial processing and intensive agricultural operations, which can pollute feed and water. It was noted that most of the wastes produced by nearby factories in Al Fallujah city are either strewn around the highways or use as landfills while the sewage is used for irrigation, while sheep graze on the edges of runways and other sites that might have been contaminated with hazardous materials. EF (enrichment factor) According to the findings, the concentration of Cd in the soils of Al Fallujah city was exceptionally high, and according to sediment content criteria, the soils in Fallujah city were contaminated with high concentrations of Cd, revealing the influence of human and industrial activity on Cd accumulation in the soil. Cadmium levels in liver, kidney and muscle from 216 samples for 72 slaughtered sheep in slaughterhouse at Al Fallujah city, Iraq during the period of October to December 2020 were analyzed. The values of Cadmium detected in sheep liver, kidney and muscle samples were 0.138± 0.020, 0.432 ± 0.090, and 0.037± 0.003 mg/kg, respectively which refer to high concentration of Cd in sheep viscera also This study demonstrates the need of assessing and monitoring cadmium and other hazardous metal levels in meals in order to create population-safe intake values.

(Invited) Microfabricated Chemical Sensors for Industrial, Health and Environmental Monitoring

There remains a strong need for miniaturized, low-power gas sensors that can be deployed in wireless applications for improved environmental protection, for public health, and for safe and efficient operation of many industrial processes. Leveraging the microfabrication technologies and utilizing innovations in high surface area nanomaterials are providing exciting opportunities towards low-power microchemical sensors. In this presentation, I will discuss one of the approaches we have been pursuing, which is based on a highly efficient microheater platform with fast response and recovery times. The design, fabrication and characteristics of this platform will be presented. I will also discuss the use of two-dimensional materials for gas sensing applications since they offer the highest possible surface area for gas interaction, leading to high sensitivity. Assembling these two-dimensional materials into three-dimensional assemblies, such as aerogels, provides a low-density material with large number of interconnected pores that increases the surface area available in a given footprint while maintaining the properties of the few-layer sheets. By integrating these nanomaterials with our microheater platform, we have achieved fast and sensitive detection of several health and environmental pollutants. The talk will end with current limitations and plans for future directions.

Decentralized PCA modeling based on relevance and redundancy variable selection and its application to large-scale dynamic process monitoring

In order to ensure the long-term stable operation of a large-scale industrial process, it is necessary to detect and solve the minor abnormal conditions in time. However, the large-scale industrial process contains a large number of complex related process variables, some of which are redundant for abnormal condition detection. To solve this problem, a new decentralized PCA modeling method based on relevance and redundancy variable selection (RRVS-DPCA) is presented. First, considering the complex dynamic relation of process variables, a variable selection strategy based on relevance and redundancy (RRVS) is designed to select variables that carried the most profitable information from different temporal dimensions for each key process variables, so the optimal variable sub-block for each individual key process variables can be obtained. Then, for each sub-block, a corresponding sub-PCA monitoring model is established. The sub-blocks’ monitoring results are combined to form a probability statistical indicator through a Bayesian inference. Finally, the weighed contribution plot method is proposed to find the root cause of a fault. The proposed method is compared with several state-of-the-art process monitoring methods on a numerical example and the Tennessee Eastman benchmark process. The comparison results illustrate the feasibility and effectiveness of the proposed monitoring scheme.

Application of hydrodynamic lubrication in discrete element method (DEM) simulations of wet bead milling chambers

Fine grinding in wet stirred media mills is a key operation in numerous industrial processes. During wet milling, grinding beads (typically millimetre scale) collide with each other under rapid agitation. These collisions transfer energy into the surrounding medium, a feed stream comprising a premixed slurry (a suspension of micron-sized particles), leading to a reduced particle size. Numerical simulation provides an opportunity to gain mechanistic insight by modelling the influences of grinding media, agitator speed and slurry rheology on the breakage. The multiphase nature of wet milling necessitates models that account for both the solid grinding beads and the slurry feed. Established fluid-particle coupled methods are in principle capable of doing this, but the need to resolve the large hydrodynamic lubrication forces between near-contacting beads demands extremely high fluid field resolution, leading to proliferating simulation size and cost. To address this challenge, a Discrete Element Method (DEM) simulation for a lab scale mill is presented in which the bead-bead interaction includes pairwise hydrodynamic lubrication in addition to frictional contact forces. The model accounts for multiphase effects in a simplified yet physically well-motivated way, circumventing the computational cost of a fully fluid-resolved model. A systematic model calibration against the laboratory mill is presented, and thereafter the model provides a good estimation for the empirical power draw across the full range of rotation speeds and considered feed viscosities. The differences in energy dissipation modes and the local distribution of collisions along the mill chamber are examined for the extreme viscosity condition, revealing that most energy dissipation is due to inter-particle forces acting tangentially (shearing and rolling) while highly energetic collisions (impact and torsion), relying on free flows and inter-bead mobility, are relatively unimportant.