Real-time Plant Data Research Articles

Machine learning applications on IoT data in manufacturing operations and their interpretability implications: A systematic literature review

Industry 4.0 has transformed manufacturing with real-time plant data collection across operations and effective analysis is crucial to unlock the full potential of Internet-of-Things (IoT) sensor data, integrating IoT with Artificial Intelligence (AI) techniques, such as Machine Learning (ML) and Deep Learning (DL). They can provide powerful predictions but anticipating issues is not enough. Manufacturing companies must prioritize avoiding inefficiencies, thereby developing improvement strategies from an Operational Excellence perspective. Here, the interpretability dimension of AI-based models could support a complete understanding of the reasons behind the outcomes, making ML and DL models transparent, and allowing the identification of the causal linkages between the inputs and outputs of the system. Within this context, this study aims first to deliver a comprehensive overview of the existing applications of Advanced Analytics techniques leveraging IoT data in manufacturing environments to then analyze their interpretability implications, referring to the interpretability as the description of the link between the independent and dependent variables in a way that is understandable to humans. Different gaps in terms of lack of full data enhancement are highlighted, providing directions for future research.

Architecting an Open-Source IIoT Framework for Real-Time Control and Monitoring in the Bioleaching Industry

Electronic waste (e-waste) contains toxic elements causing an important impact on environmental and human health. However, the presence of valuable metals, such as copper or gold, among others, make recycling a necessity for obtaining an alternative source of raw materials. Conventional metal recovery methods are environmentally unsound, prompting the exploration of greener alternatives like bioleaching, which utilizes the activity of microorganisms for a more sustainable recovery. However, the mechanisms involved in the process and the conditions to optimize the metabolic paths are still not completely known. Monitorization and automatization of the different stages composing the global process are crucial for advancing in the implementation of this novel technology at an industrial scale. For the first time, an open-source industrial IoT system is designed to enhance and regulate bioleaching by implementing real-time monitoring and control within the plant’s infrastructure. This system includes an Android app that displays real-time plant data from sensors and a robust server featuring a flexible application programming interface (API) for future applications. The app caters to specific needs such as remote sensor reading, actuator control, and real-time bioleaching alerts, ensuring secure access and proactive event management. By utilizing collected data, it minimizes downtime, equipment failures, and supply chain disruptions. The server maintains seamless communication with the plant controller, enabling efficient pump activation and sensor data transmission. A telegram bot demonstrates the API’s flexibility by forwarding plant alerts to users. During validation with concurrent remote user access, the application demonstrated its ability to prevent irreversible plant failures through an advanced alarm system. Ultimately, this IIoT system amplifies plant performance, safety, and efficiency by optimizing processes and decision-making capabilities. It emerges as a pivotal open-source tool, securing remote oversight and management of large-scale bioleaching plants, promising adaptability for future enhancements.

Evaluation and Implementation of WISE-4051-AE Communications in Process Level Laboratory Module

This research discusses the design and implementation of wireless communication using WISE-4051-AE for optimizing process-level practical modules. The WISE-4051-AE wireless device is an integrated IoT (Internet of Things) device that combines data acquisition, processing, and control functionalities. Continuing previous research, this study focuses on creating practical modules for process control, including flow meter calibration, cascade control, feedforward control, and ratio control. By utilizing the WISE-4051-AE wireless device in a plant setting, the processed data can be accessed through laptops, personal computers, and other devices that support communication with the WISE-4051-AE device. To get the data, a device must be connected to the WISE-4051-AE wireless device. By employing the WISE-4051-AE wireless device for data transmission, users of the process-level practical modules can receive real-time plant data, regardless of whether the modules are indoors or outdoors, without encountering any data alteration during the transmission process. In this research, the WISE-4051-AE wireless device has a maximum range limitation. As long as the user's device remains within the signal range of the WISE-4051-AE device, the transmitted data will not undergo any changes. The signal range was tested under two conditions: without signal obstacles and with signal obstacles. The results of the signal range testing revealed a maximum range of 90 meters without obstacles and 40 meters with obstacles.

Fault detection and isolation of multi-variate time series data using spectral weighted graph auto-encoders

Fault or anomaly detection is one of the key problems faced by the chemical process industry for achieving safe and reliable operation. In this study, a novel methodology, spectral weighted graph autoencoder (SWGAE) is proposed, wherein, the problem of anomaly detection is addressed with the help of graphs. The proposed approach entails the following key steps. Firstly, constructing a spectral weighted graph, where each time step of a process variable in the multivariate time series dataset is modelled as a node in an appropriately tuned moving window. Subsequently, we propose to monitor the weights of the edges between two nodes that make a connection. The faulty instances are identified based on the discrepancy in the weight pattern compared to normal operating data. To this end, once the weights are determined, they are fed to the auto-encoder network, where reconstruction loss is calculated, and faults are identified if the reconstruction loss exceeds a threshold. Further, to make the proposed approach comprehensive, a fault isolation methodology is also proposed to identify the faulty nodes once the faulty variables are identified. The proposed approach is validated using Tennessee-Eastman benchmark data and pressurized heavy water nuclear reactor real-time plant data. The results indicate that the SWGAE method, when compared to the other state-of-the-art methods, yielded more accurate results in correctly detecting faulty nodes and isolating them.

Design modification in an industrial multistage orifice to avoid cavitation using CFD simulation

BackgroundMulti-stage orifices (MSOs) are used in many industries such as oil refineries, nuclear power plants, fertilizer plants and chemical plants for pressure reduction applications. This pressure reduction often leads to an outbreak of vapors from liquid water resulting in cavitation. Cavitation imposes many problems, including loss of efficiency, erosion, and vibrations or noise. In order to mitigate the effects of cavitation on the system, it is essential to thoroughly understand the occurring phenomenon. The main objective of this study is to propose a design modification for an industrial multi-stage orifice that inhibits the onset of cavitation. MethodComputational fluid dynamics (CFD) analysis was made to simulate an actual three dimensional setup of a six-stage orifice in which cavitation occurs after the last stage of the orifice. CFD model was validated by comparing the real-time plant data with actual operating conditions. Significant FindingsVarious modifications in the geometry were tested with a focus on the orientation of the holes in orifices to eliminate the cavitation problem associated with the existing geometry. Multi-holed geometry having four holes in each orifice plate arranged at alternate positions was observed to be the best performing geometry without any cavitation effects under the same operating conditions.

Mixed-Control Kinetic Model of Magnesia-Based Flux Dissolution in BOF Steelmaking Slags: Part 2 Model Validation With Real-Time Plant Data

Mixed-Control Kinetic Model of Magnesia-Based Flux Dissolution in BOF Steelmaking Slags: Part 2 Model Validation With Real-Time Plant Data

Discrete Simulation Model of Industrial Natural Gas Primary Reformer in Ammonia Production and Related Evaluation of the Catalyst Performance

The catalytic steam reforming process of natural gas consumes up to approximately 60% of overall energy used in ammonia production. The optimisation of the reforming catalyst performance can significantly improve the operation of the whole ammonia plant. An online model uses actual process parameters to optimise and reconcile the data of primary reforming products with possibility to predict the catalyst performance. The model uses a combination of commercial simulator and open-source code based on scripts and functions in the form of m-files to calculate various physical properties of reacting gases. The optimisation of steady-state flowsheet, based on real-time plant data from the distributed control system (DCS), is essential for the application of the model at the industrial level. The simplicity of the calculation method used by the model provides the fundamental basis for industrial application in the frame of digitalisation initiative. The principal aim of the optimisation procedure is to change the working curve for methane regarding its equilibrium curve as well as methane outlet molar concentration. This is the critical process parameter in reforming catalyst operation. An industrial top fired primary reformer unit based on Kellogg Inc technology design served for the validation of the model. Calculation procedure is used for continuous online evaluation of the most commercially available primary reformer catalysts. Based on the conducted evaluation, the model can indicate possible recommendations which can mitigate marginal performance and prolong reformer catalyst lifetime.

A digital twin approach to system-level fault detection and diagnosis for improved equipment health monitoring

Automating the task of fault detection and diagnosis is crucial in the effort to reduce the operation and maintenance cost in the nuclear industry. This paper describes a physics-based approach for system-level diagnosis in thermal–hydraulic systems in nuclear power plants. The inclusion of physics information allows for the creation of virtual sensors, which provide improved fault diagnosis capability. The physics information also serves to better constrain diagnostic solutions to the physical domain. As a demonstration, various test cases for fault diagnosis in a high-pressure feedwater system were considered. The use of virtual sensors allows constructing performance models for two first-point feedwater heaters which would not have been possible otherwise due to the limited sensor set. Real-time plant data provided by a utility partner were used to assess the diagnostic approach. The detection of an abnormal event immediate after a plant startup pointed to faulty behaviors in the two first-point feedwater heaters. This double-blind fault diagnosis was subsequently confirmed by the plant operator. In addition, several simulated sensor fault events demonstrated the capability of our algorithms in detecting and discriminating sensor faults.

Performance optimization of multiple stage evaporator using interior-point method and metaheuristic approaches in environment of real-time plant complexities

ABSTRACT This paper aims to maximize the energy efficiency of Multiple Stage Evaporator (MSE) to increase the solid concentration of the black liquor in the Kraft recovery process. The incorporation of different energy reduction schemes ensures a considerable magnification in energy efficiency, which is realized by developing a set of nonlinear energy models of the MSE. These models are optimized to attain the optimum process parameters that estimate the optimum energy efficiency in Steam Economic (SE) and Steam Consumption (SC). The formulated optimization problems are efficiently solved using three novel approaches, Interior Point Method, Particle Swarm Optimization, and Artificial Electric Field Algorithm. With these algorithms’ advent, the obtained results are approaching efficient outcomes (SE increased by 77.58% and SC decreased by 26.99% compared to real-time plant data). Further, the present work investigates the influence of Boiling Point Elevation and Fouling’s effect on energy efficiency, which causes a reduction in SE (7.189). Finally, this study extends up to utilizing the waste steam as a heat source, collected from the condensate to accelerate the evaporation process, enhancing the SE by 2.94%.

Development and implementation of key performance indicators for aggregate production using dynamic simulation

Continuous process improvements are needed to maintain and run an industrial operation at a profitable state. An aggregates production plant consists of multiple process operations such as blasting, primary crushing, followed by secondary and tertiary crushing stages with an intermediate screening of products. Each of these processes can be operated with multiple objectives in mind of operators and plant managers. These objectives can be defined by varying terms like generating required throughput of the plant, maintaining equipment’s health, meeting customers’ demands, etc. The use of the term key performance indicators is recurrent in industry to formalise and represent these objectives of operation. Currently, the KPIs defined by the ISO 22400 standards are widespread for continuous improvements in the manufacturing industry and they are viewed as a support tool to measure improvements. The scope of this paper is to calculate relevant KPIs for an aggregates production plant using dynamic simulations. Further, the KPIs are implemented in a three-stage aggregate production plant using both real-time plant data and dynamic process simulation. The KPIs developed are useful for operators and plant managers to make decisions. The results show the relationship and dependencies of various equipment and process KPIs. The dynamic simulation has potential to be used as an exploration tool to identify the opportunities of improvement in aggregates processing using KPIs as a measure. The KPIs presented in the paper are based on ISO 22400 standard and have potential to be extended to similar processing operations such as coarse and fine comminution for minerals processing. Apart from the diagnostics application, the KPIs implemented in the dynamic simulation platform can be used to explore and optimize a crushing plant’s design and operations.

Solidification behavior of binary eutectic phase change material in a vertical finned thermal storage system dispersed with graphene nano-plates

Thermal energy storage systems (TES) find their extensive application in industrial and solar-powered thermal systems due to their high energy storage density and ability to deliver heat at isothermal conditions. But, the lower thermal conductivity of phase change materials (PCM) reduces the overall rate of heat transfer. In the present work, solidification behavior in a medium temperature (160–200 °C) finned TES with varying proportion of Graphene nano plates (GNP) is presented. The effect of anisotropy, aspect ratio, concentration, interfacial thermal resistance and non-linearity has been included to evaluate the overall thermal conductivity of GNP-PCM composite. The objective of the present study is to enhance the solidification process of a binary eutectic salt LiNO3-KCl in an optimized finned TES dispersed with the different volume fraction of GNP. The detailed experimentation process starts with the preparation of binary eutectic mixture, followed by dynamic Differential scanning calorimetry analyses along with rheology testing. Empirical correlations have been developed using quadratic and cubic polynomials for viscosity and specific heat respectively. Several numerical models are analyzed to study the solidification enhancement of binary eutectic PCM using the real-time plant data of a high-temperature solar absorption chiller. Effects of Stefan and Reynolds number on the thermal performance of storage system has also been studied. Reducing the Stefan number and increasing the Reynolds number of heat transfer fluid (HTF) results in enhancing the rate of solidification process. It has been observed that natural convection currents enhances the rate of solidification and causes faster solidification in the upper annulus of TES. A reduction of 49% in the solidification time has been observed with finned TES dispersed with 5% GNP as compared to pure binary eutectic PCM.

The Power Plant Operating Data Based on Real-time Digital Filtration Technology

Real-time monitoring of the data of the thermal power plant was the basis of accurate analyzing thermal economy and accurate reconstruction of the operating state. Due to noise interference was inevitable; we need real-time monitoring data filtering to get accurate information of the units and equipment operating data of the thermal power plant. Real-time filtering algorithm couldn’t be used to correct the current data with future data. Compared with traditional filtering algorithm, there were a lot of constraints. First-order lag filtering method and weighted recursive average filtering method could be used for real-time filtering. This paper analyzes the characteristics of the two filtering methods and applications for real-time processing of the positive spin simulation data, and the thermal power plant operating data. The analysis was revealed that the weighted recursive average filtering method applied to the simulation and real-time plant data filtering achieved very good results.

Smart manufacturing and energy systems

While many U.S. manufacturing operations utilize optimization for individual unit processes, smart manufacturing (SM) systems that integrate manufacturing intelligence in real time across an entire production operation are not pervasive in industry. A vendor-agnostic SM platform is under development that integrates information technology, models, and simulations driven by real-time plant data and performance metrics. By utilizing existing process control and automation systems, manufacturing organizations can manage systems at a much lower cost, optimizing process knowledge and improving energy productivity. Three case studies are presented: steam methane reforming to make hydrogen, optimization of a heat treatment furnace for metals processing, and a fuel cell system, all of which utilize high fidelity models as a starting point for optimization and control. The Smart Manufacturing Leadership Coalition has led the national effort in SM, and the recently established National Manufacturing Innovation Institute funded by DOE, private industry, and state governments will be described.

Mathematical model for integrated coal fired thermal boiler using physical laws

The boiler of coal-fired thermal power plant consists of integrating process units such as the furnace, economizer, drum, and superheater. There is a lack of proper dynamic model derived from mass and energy balances of individual units integrated together where dynamics of material and energy flow are not understood properly supporting the larger footprints and environmental impact of the specific industry. This paper considers formulation of simple mathematical models for integrated boiler units based on first principle laws (excluding furnace). The derived equations are validated with the real-time plant data from 210 MW coal-fired thermal power plant. Though temperature and pressures are convertible at equilibrium through Clausius-Clapeyron equation, the model supports for level, temperature and pressure measurements and predictions. The effects of change in parameters of the boiler are studied and discussed in detail. The model can be used for further process control studies.

Process-structure-microstructure Relationship in Hot Strip Rolling of Steels Using Statistical Data Mining

Mathematical models have been widely used for prediction of microstructure and mechanical properties in hot rolling of strip. To accurately predict these characteristics, it is necessary to create models that can replicate thermomechanical state of material and its evolution during processing. This paper presents development of a hybrid model that uses mills setting and real time plant data such as chemical composition; forces and temperatures; and integrates them with empirical relationships of material evolution to predict quality attributes. This information is combined with non-linear statistical data mining models to create online tool that predicts properties of individual coil. Case study from Steel Plant is presented that illustrates implementation, calibration and validation of this model across different materials grades.

The Research of Power plant Operating Data Based on Real-time Digital filtration technology

Real-time monitoring of the data of the thermal power plant is the basis of accurate analyzing thermal economy and accurate reconstruction of the operating state. Due to noise is inevitable, we need real-time monitoring data filtering to get accurate information of units and equipment in the operating data of thermal power plant. Real-time filtering algorithm can’t be used to correct the current data with future data. Compared with traditional filtering algorithm, there are a lot of constraints. First-order lag filtering method and weighted recursive average filtering method can be used for real-time filtering. This paper analyzes the characteristics of the two filtering methods and applications for real-time processing of the positive spin simulation data, and the thermal power plant operating data. The analysis revealed that the weighted recursive average filtering method applied to the simulation and real-time plant data filtering achieved very good results. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.3468

A drinking water treatment plant simulator using real-time plant data for enhanced operator training and model evaluation

Waterspot is a plant-wide drinking water treatment simulator that has been developed in a 3 year research project. The incorporation of real-time and historical data of on-line water quality measurements, flow measurements, process data and water quality laboratory measurements in the simulator improves the acceptance of model output by technologists, ensures a more realistic training of operators and shows the potential of detecting deviations from the expected water quality and control of the plant. The optimal position of the simulator in the automation architecture was found to be on the process information management system (PIMS) network.

An On-Line Hierarchical Decomposition Based Bayesian Model for Quality Prediction during Hot Strip Rolling

Mathematical models have been widely used for the prediction of the microstructure and mechanical properties in hot rolling of strip. However, their accuracy is insufficient for quality control purposes. To accurately predict these characteristics, it is necessary to create models which can replicate the thermomechanical state of the material and its evolution during processing. In addition, these models should be able to capture the uncertainties introduced in the processing by the dynamics of the casting process and the subsequent rolling. These uncertainties lead to considerable variations in the material and mechanical state of the rolled strip. This paper presents the development of a hybrid model (MICROL) which uses the mills setting and the real time plant data such as chemical composition; forces and temperatures; and integrates them a Bayesian format to predict the desired quality attributes as well as microstructural features. This information is combined into Bayesian Hierarchical models to create an on-line tool that predicts the properties of each individual rolled coil, as well as provide information on the batch-to-batch and heat-to-heat variations. Case study from a steel Plant is presented which illustrates the implementation, calibration and validation of this model across different materials grades. Model results are found to be within the 5% tolerance of the measured values for many steel grades and rolling conditions.

Evaluation of flotation performance using variance spectrum analysis

The performance of Denver sub-A flotation cells in rougher and 1st cleaner service in Inco's Matte Separation plant was studied using the technique of Variance Spectrum Analysis. The procedure involves fitting a Fourier Series to real-time plant data and filtering out the low frequency disturbances on the assumption that performance would be enhanced with improved equipment installation — in this case Denver DR flotation cells. The resulting series is then recalculated to estimate the response of the process under improved control. In the case under study, recovery of nickel to the high grade nickel sulfide product was projected to increase by 5 % as a result of implementing the equipment upgrade in the rougher circuit. In addition, it was projected that the quality specification target of a secondary nickel sulfide product could be met simultaneously with that of the copper sulfide product, by upgrading the 1st cleaner cells. These targets had previously been difficult to achieve concurrently due to the strong inverse correlation between them. Based on this study an equipment upgrade was carried out. Subsequent operation of both circuits confirmed the projected results and the validity of the method as a basis for justifying capital investment based on improved metallurgical performance.