ABSTRACT
Purpose: The project aims to leverage Business Intelligence integrated with dynamic simulations in the context of Industry 4.0, focusing on the implementation of a system for the dynamic analysis of a chemical reaction process. The system is designed to optimize data analysis and decision-making processes in large-scale industrial settings.
Theoretical Framework: In the evolving landscape of Industry 4.0, the effective management of industrial data and processes is paramount. Traditional methodologies often overlook the potential of integrating Business Intelligence with dynamic simulations. This project proposes a novel approach, combining these elements to enhance process oversight and decision-making efficacy.
Method: Utilizing PI System AVEVA/OSisoft for structured data organization, the project implements a Business Intelligence framework applied to the dynamic simulation of a chemical reaction process. This method focuses on creating a harmonious integration of data analysis tools with real-time process simulations, aiming to improve operational efficiency and adaptability.
Results and Conclusion: The implementation of this integrated system in a simulated industrial environment demonstrated a notable improvement in process analysis and decision-making efficiency. This indicates a significant advancement in the application of Business Intelligence in industrial operations, particularly in dynamic process simulations.
Research Implications: This study underscores the importance of advanced data analysis techniques in modern industrial operations. The results suggest a substantial shift towards more data-driven, efficient, and adaptable process management strategies in Industry 4.0.
Originality/Value: This research highlights the innovative application of Business Intelligence in conjunction with dynamic simulations for industrial processes. It showcases a pioneering approach in enhancing data analysis and decision-making capabilities in the context of Industry 4.0.
Keywords: Business Intelligence, Industry 4.0, Dynamic Simulation, Chemical Process, Data Analysis.
RESUMO
Objetivo: O projeto visa aproveitar a Inteligencia de Negocios integrada a simulaçôes dinámicas no contexto da Industria 4.0, focando na implementaçâo de um sistema para a análise dinámica de um processo de reaçâo química. O sistema é projetado para otimizar a análise de dados e os processos de tomada de decisao em ambientes industriáis de grande escala.
Referencial Teórico: No cenário em evoluçâo da Industria 4.0, a gest&acaron;o eficaz de dados e processos industriáis é primordial. Metodologías tradicionais muitas vezeš negligenciam o potencial da integraçào da Inteligencia de Negocios com simulaçôes dinámicas. Este projeto propôe uma abordagem inovadora, combinando esses elementos para aprimorar a supervis&acaron;o de processos e a eficacia na tomada de decisao.
Método: Utilizando o PI System AVEVA/OSIsoft para organizaçâo estruturada de dados, o projeto implementa um framework de Inteligencia de Negocios aplicado à simulaçâo dinámica de um processo de reaçâo química. Este método foca em criar uma integraçào harmoniosa de ferramentas de análise de dados com simulaçôes de processos em tempo real, visando melhorar a eficiencia operációnál e a adaptabilidade.
Resultados e Conclusâo: A implementaçâo deste sistema integrado em um ambiente industrial simulado demonstrou uma melhoria notável na análise de processos e na eficiencia da tomada de decisôes. Isso indica um avanço significativo na aplicaçâo da Inteligencia de Negocios em operaçôes industriáis, particularmente em simulaçôes de processos dinámicos.
Implicaçöes da Pesquisa: Este estudo sublinha a importancia de técnicas avançadas de análise de dados em operaçôes industriáis modernas. Os resultados sugcrem uma mudança substancial em direçâo a estrategias de gestâo de processos mais orientadas a dados, eficientes e adaptáveis na Industria 4.0.
Originalidade/Valor: Esta pesquisa destaca a aplicaçâo inovadora da Inteligencia de Negocios em conjunto com simulaçôes dinámicas para processos industriáis. Apresenta urna abordagem pioneira no aprimor amento das capacidades de análise de dados e tomada de decisôes no contexto da Industria 4.0.
Palavras-chave: Inteligencia de Negocios, Industria 4.0, Simulaçâo Dinámica, Processo Químico, Análise de Dados.
1 INTRODUCTION
The industrial landscape has been continually shaped by revolutionary periods, each introducing transformative changes. The current period, often referred to as the fourth industrial revolution, is marked by the incorporation of advanced technological paradigms, including cyber-physical systems, the Internet of Things (loT), and Big Data. These technologies are fundamentally altering the way industrial operations are conducted, emphasizing smart automation and data-driven decision-making. This shift is not merely a technological upgrade, but a comprehensive redefinition of how industrial processes are managed and optimized.
In this era of Industry 4.0, a significant volume of process data is generated and stored daily, presenting unprecedented opportunities for insight extraction and business enhancement. The crucial role of structured data organization and advanced analytical computations cannot be overstated. These tools enable the formulation of coherent decisions, significantly reducing errors and inefficiencies in industrial operations. Such advancements highlight the transition from traditional analog control systems to sophisticated digital and integrated supervisory systems, positioning tools like the PI System AVEVA/OSIsoft at the forefront of data management in the chemical industry.
The backbone of this industrial evolution is a robust data communication infrastructure. This infrastructure is crucial for the precise transport and delivery of data, ensuring seamless integration across various operational platforms. Automation and control networks, predominantly utilizing Open Platform Communications (OPC) protocols, facilitate this integration. OPC standardizes data communication, enabling diverse applications to interact and interpret data uniformly, regardless of the source's formatting standards. This unification is pivotal in maintaining the integrity and consistency of process control data across different industrial systems. This paper explores the integration of such systems within the industrial context, underlining their potential to revolutionize traditional manufacturing paradigms.
2 THEORETICAL REFERENCE
The world is currently experiencing the fourth industrial revolution, characterized by the implementation of a suite of technologies founded on the concepts and interactions among cyber-physical systems, the Internet of Things, and Big Data. These technologies aim to facilitate decision-making processes for operators in smart factories/units (Schwab, 2017). Such amount of process data is generated daily, stored in specialized databases. This indicates a significant potential to extract insights for business improvement from such information. With structured organization and the implementation of analytical computations, more coherent decisions can be readily made, thereby reducing the margin of error. This smart industry, named as Industry 4.0, refers to the fourth industrial revolution: the data revolution. In this context, chemical engineers play a crucial role (Kestering et al., 2023).
Behind every process, a data communication structure is essential, responsible for transporting and delivering data as precisely as possible. Generally, automation and control networks use various communication protocols, with OPC being the most common. Such a communication protocol allows external applications to interact through a simplified interface, enabling them to receive data in a uniform format, even if the data source operates with a different standard/formatting and communication, thus unifying the communication standard of process control data.
It is important to note that this advanced manufacturing structure exists in the industrial environment and can be used to support decision-making. However, such decisions should be made at three levels: operational, tactical, and strategic, as seen in the enterprise pyramid in Figure 1. This vertical hierarchy is present in most industries, and even in more horizontal structures, decision-making remains at these three levels.
Each level has specific functions and is usually interrelated. The operational level consists of workers and engineers on the process unit floor, directly related to production engineering and responsible for most of the unit's labor. The tactical level comprises sector managers, requiring more information for decision-making and topics related to process engineering, making integrated information systems fundamental in the organizational structure. The strategic level includes directors, presidents, and CEOs, responsible for long-term decision-making. As this is a critical sector in the process, where errors can lead to significant losses, strategic managers need not only support for their decisions but also a foundation for them, which is aided by the implementation of a Business Intelligence (BI) process.
BI consists of a set of tools that enable the collection, organization, and analysis of data from various processes and projects. In industry, this means relying on software that can "sift" information at all stages of the production chain, from customer orders and raw materials to production and distribution. This provides managers with a broader view of production, allowing more efficient actions for quality management without constantly involving IT (Information Technology). With technological evolution, there is now the possibility of expanding the understanding of processes through the analysis of results from process simulations, which can be stationary and/or dynamic. Such a tool is becoming essential from both economic and safety viewpoints. Its use is justified as it does not require substantial physical or economic investments and, most importantly, does not pose risks to operational safety.
Once well-developed, the model can be statistically validated, and the results will represent the unit's behavior. At this point, various operational conditions can be tested, as well as control methodologies, optimization, etc., in a quick and practical manner (Stephanopoulos, 1984; Roffel and Betlem, 2006; Edgar et al., 2001). Various software programs are used for dynamic process analysis, from general-purpose programming languages such as VBA, MATLAB, Fortran, etc. (Finlayson, 2006; Attaway, 2009) to simulation platforms like Aspen Plus Dynamics AspenTech and DynSim AVEVA. Therefore, this project aims to implement a data flow structure between simulation software for developing asset analysis systems that simulate a real operational environment. These systems could eventually be used by actual units, favoring decision-making at various levels of a unit.
3 METHODOLOGY
The integration of many simulation software tools are fundamental in the field of chemical process modeling. Among these, Aspen Dynamics is a market leader, widely recognized for its comprehensive capabilities. The selection of this particular software is primarily driven by its efficient data flow handling, which is decisive for the successful implementation of any process simulation project. In the area of process data management, the PI System AVEVA/OSISoft emerges as a pivotal tool, enabling the capture, storage, structuring, and visualization of data, as presented by Figure 2. This platform facilitates the comprehensive structuring of the process into elements and attributes, including the incorporation of market-related information (Kestering et al., 2023).
For effective communication between the dynamic process and the data acquisition system, the industrial standard OPC (Open Platform Communications) protocol is chosen. This protocol's widespread industrial use makes it an ideal choice for ensuring seamless data transfer to the PI System trough PI ICU (Interface Configuration Utility). The project further involves the creation of HMI (human-machine interfaces), allowing for an enhanced visualization and interaction with the process data, thereby simplifying the management and analysis of complex process information.
The development phase of the project commenced with an in-depth familiarization with the software tools involved, to a better proficiency on the use of the platforms. The primary focus was on implementing a Business Intelligence system, anchored by a simulation developed in Aspen Plus Dynamics/AspenTech. This simulation, based on data derived from Fogler (2009), serves as the foundation for the system's implementation. The project involved analyzing and validating design conditions through direct comparisons between the results obtained from the literature example and the simulation. To facilitate a comprehensive decision-making process, the example was implemented across three types of reactors: a CSTR (Continuous Stirred Tank Reactor), a PFR (Plug Flow Reactor), and a stoichiometric reactor. The simulation data was then exported to Aspen Plus Dynamics and configured for OPC communication. Simultaneously with the simulation development, the installation and utilization of the PI System started. For this phase online courses are indicated:
1. Configuring a Simple PI System
2. Building PI System Assets and Analytics with PI AF
3. Analyzing PI System Data
4. Creating Basic Reports with PI DataLink
5. Visualizing PI System Data with PI Vision
Such kind of personal improvement is a key component for developing the communication structure between Aspen Plus Dynamics and the PI Data Archive. These courses facilitated the familiarization with visualization software and enabled the creation of user-friendly human-machine interfaces. This comprehensive approach to software learning and integration ensures a robust foundation for the project, allowing for efficient data analysis and system operation in the complex field of chemical process modeling.
4 RESULTS AND DISCUSSION
An important step on this research is related to the focus on the practical application of Business Intelligence tools in the chemical engineering area. For this is need the exploration of process modeling, thus the Hydrodealkylation of Mesitylene has been chosen. Such a process has been extensively researched and documented by Fogler in 1999. The choice of this particular process for the initial model is grounded in its familiarity and the extensive body of work surrounding it, exemplifing the complex dynamics of chemical reactions, involving two sequential stages: the transformation of mesitylene into m-xylene, followed by the conversion of m-xylene into toluene.
The successful implementation of this process in Aspen Plus marked a significant goal, achieving the targeted conversion rates essential for the project. This accomplishment covered the way for the extraction of critical steady-state data, subsequently used to enrich the dynamic simulation on Aspen Plus Dynamics. A key aspect of this simulation was the incorporation of realistic operational elements to closely mimic an real process plant. This was achieved by integrating three compressors to regulate reactor pressures and introducing a simulated noise in the feed stream valve. These enhancements not only add a layer of realism to the simulation but also provide valuable insights into the practical challenges encountered in industrial processes. As depicted in Figure 3, this approach bridges the gap between theoretical models and real-world applications, setting the stage for a deeper exploration of the applications and implications of Business Intelligence in chemical engineering.
To keep the process on the desired steady state during the dynamic simulation, proportional-integral controllers have been implemented to ensuring process stability. For this:
1. Two Proportional-Integral (PI) controllers (Pressure and Temperature) to the Continuous Stirred Tank Reactor (CSTR)
2. One PI controller was employed to the Plug Flow Reactor (PFR) for pressure control.
After the validation of the process, the next step was the establishment of a communication structure between the simulator and the PI System. The chosen communication protocol was OPC, because it is a standard protocol in the industry facilitating interaction between software and industrial plants via a client/server architecture.
On this work, this protocol played a pivotal role in standardizing the communication, bridging the gap between theoretical models and practical applications. For effective communication, process variables of interest were converted into tags, enabling the transmission of data between different software platforms through strings. Figure 4 showcases the selected tags in the simulation, now seamlessly connected to the OPC server. Such a communication is possible due to a native tool available at Aspen Dynamics named "On Line Links". This illustration provides a visual representation of the simulation's complexity by meticulous process of linking theoretical models with practical.
Some process variables, important for the dynamic simulation analysis, were systematically stored in the OPC server. Such a server is responsible on the standardizing the communication between diverse software platforms, acting as a link in the data exchange chain. The use of an OPC server give a more robust, reliable, and seamless communication across different platforms, a crucial aspect in process simulation and control area.
The MatrikonOPC Server for Simulation was chosen due to its compatibility and effectiveness in handling simulation data. Within this server environment, a comprehensive set of tags corresponding to the selected process variables was created. These tags form the backbone of the data exchange, connecting the dynamic simulation with the PI System. The creation and integration of these tags into the simulation and the PI System are steps that ensure the data flow in an uninterrupted and efficient way. Figure 5, depicting the structured tags, offers more than a visual representation; it serves as a detailed map of the data linkage within the project.
With the entire OPC communication structure established, using a communication interface (PI Interface Configurator Utility), these tags were sent to the PI Server (Data Archive), where they were stored and managed according to their purpose. The significant advantage of using the PI System is that it allows for greater structuring in the hierarchy of tags. It enables the organization of tags according to their equipment, unit of measure, and type, thereby allowing the addition of contextual information to the data.
Figure 6 illustrates the communication diagram between Aspen Plus Dynamics and the PI System. In this schema, the initial variable is acquired from the industrial plant simulation via the PI Interface Configurator Utility and then stored in the PI Server. Here, the data undergo treatment and management before being dispatched to client applications. With a functional communication structure in place, the process flowchart was efficiently organized within the PI Asset Framework (PI AF), utilizing equipment templates. This was particularly effective as identical pieces of equipment shared the same variables of interest. Moreover, up-to-date market information was integrated to enhance managerial decision-making at the plant level, as depicted in Figure 7. The PI AF also facilitated the implementation of calculation routines, enabling dynamic computation of the conversion rates for each reactor. This approach not only streamlined the data management process but also allowed for more agile and informed decision-making based on real-time data analysis.
To improve the visualization of process data and information derived from implementation, multiple dashboards were developed using the PI System visualization tool: PI Vision. This allowed for comprehensive analysis of the plant's historical data via web browsers, as well as the evaluation of key parameters of selected reactors and compressors, including moments of peak and minimal conversion. Incorporating PI Vision into the research framework significantly contributed to the project's analytical depth. It facilitated a more intuitive and interactive way of examining complex data sets, enabling researchers and plant managers to swiftly navigate through various data points. The ability to visualize trends, patterns, and anomalies in real-time provided a more nuanced understanding of the plant's operational dynamics. This level of insight is invaluable for optimizing process efficiency, identifying potential areas of improvement, and making informed decisions based on comprehensive data analysis. The integration of such advanced visualization tools in chemical engineering research underlines the growing importance of data-driven approaches in industrial process management and optimization. Figure 8 displays the main screen of the process, showcasing the interface's capabilities.
When selecting the "R_101 " display, a new screen opens, revealing detailed information specific to the reactor. This interface provides crucial data such as the quantities of products and reactants, and it also features alarm indicators that highlight risk situations that have occurred. This level of detail is instrumental for monitoring and managing the reactor's performance effectively. Both the main and reactor-specific screens are designed to be used for all three reactors, with the capability to switch between them using the "Asset" option. This user-friendly design ensures that the same interface layout can be efficiently applied to each reactor, maintaining consistency and ease of use. The flexibility of this approach allows for quick comparisons and assessments across different reactors, streamlining the analysis process. Figure 9 below presents this reactor-specific screen, showcasing its functionality and layout.
The implementation of such tailored interfaces in PI Vision underlines the significance of adaptable visualization tools in chemical engineering research. By providing customized views that cater to specific equipment or process stages, these interfaces enhance the user's ability to monitor, analyze, and respond to process changes or anomalies. This adaptability not only improves operational efficiency but also plays a crucial role in risk management and safety assurance. Such advancements in data visualization represent a key step forward in the integration of digital tools into the field of process engineering, highlighting the importance of technology in facilitating more effective and informed decision-making.
To increase the management of both the process and data derived from dynamic simulation, a daily management report for the plant was developed using a feature of the PI System, specifically PI DataLink, which facilitates the export of data to a Microsoft Excel spreadsheet acting as a dynamic and smart spreadsheet. This functionality allows for a detailed analysis of historical daily values, encompassing cash flow, material flow, quantities of products and reactants, and the energy consumption of the plant. The capability of this setup is illustrated in Figure 10. The integration of PI DataLink into the process management infrastructure exemplifies an effective application of Business Intelligence within chemical engineering. By providing a streamlined channel for data transfer and analysis, PI DataLink enhances decision-making efficiency. The ability to easily access and analyze comprehensive data sets empowers operators and engineers to creatively engage in process improvement and optimization endeavors. Having this wealth of data at their disposal, in a format that is readily usable and adaptable, allows for a more dynamic and innovative approach to managing plant operations.
This infrastructure, connecting data from the simulated unit with the PI System components (PI ICU, PI AF, PI Vision), and now enriched with PI DataLink, stands as a potent example of Business Intelligence in action. It transforms the process of data handling from a mere operational task into a strategic asset, facilitating not just the routine management of the plant but also fostering an environment conducive to continuous improvement and innovation. The availability and versatility of data, made possible by this integrated system, are crucial for advancing process efficiency, ensuring sustainability, and driving optimization in the field of chemical engineering.
5 CONCLUSION
The implementation of the data capture structure presented in this project opens up a countless of analytical possibilities and demonstrates the versatile application of these tools in industrial projects. It significantly enhances the ability of engineers, managers, and operators to visualize various simulated and real processes, thereby bridging the gap between theoretical models and practical operations. This accessibility is crucial for informed decision-making and efficient process management.
Further, the development of calculation routines for dynamic simulations has greatly simplified the interpretation of the process's phenomenological model. This simplification enables the incorporation of various additional pieces of information into the process, enriching the overall understanding and control of the process dynamics. Such enhancements in the simulation models contribute to a more comprehensive and nuanced approach to process analysis and optimization.
The system developed is readily applicable to various real-world plants. Its adaptability lies in the ease of modifying the data source, which offers a unique advantage for training operators in simulated scenarios. This flexibility not only aids in the practical training of personnel but also ensures that the transition from a simulated environment to practical operations is smooth and effective.
Moreover, the interfaces and reports developed as part of this system facilitate an intuitive visualization of process data. The ability to access data from various mobile sources is a significant advancement, increasing the ease and flexibility with which information can be reviewed and analyzed. Importantly, the system has been designed with process safety and data security in mind. Access to data is carefully controlled, ensuring that only authorized users can view and interact with sensitive information. This consideration is paramount in maintaining the integrity and security of both the data and the process.
In conclusion, this paper has presented showcases of advanced data analysis integration and visualization tools in chemical process engineering and highlights the importance of such technologies in enhancing process understanding, efficiency, and safety, once the data-driven decision-making and process optimization are central to the industry.
6 Universidade Federal de Campina Grande, Campina Grande, Paraíba, Brazil. E-mail: [email protected] Orcid: https://orcid.org/0000-0002-2138-124X
REFERENCES
ATTAWAY, S., MatLab a Practical Approach, Elsevier, 2009.
EDGAR, T., HIMMELBLAU, D., E LASDON, L., Optimization of Chemical Processes, McGraw-Hill, 2001.
FINLAYSON, B., Introduction to Chemical Engineering Computing, John Wiley & Sons, 2006.
FOGLER, H., Elements of Chemical Reaction Engineering, Prentice Hall, 1999.
KESTERING D, AGBLEZE S, BISPO H, LIMA VF, Model predictive control of power plant cycling using Industry 4.0 infrastructure. Digital Chemical Engineering, vol. 7, 2023.
ROFFEL, B. E BETLEM, В., Process Dynamics and Control - Modeling for Control and Prediction, John Wiley & Sons, 2006.
SCHWAB, K. The Fourth Industrial Revolution. World Economic Forum, 2016.
STEPHANOPOULOS, G., Chemical Process Control: An Introduction to Theory and Practice, Prentice-Hall, 1984.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2024. This work is published under https://rgsa.emnuvens.com.br/rgsa/about/editorialPolicies#openAccessPolicy (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Objetivo: O projeto visa aproveitar a Inteligencia de Negocios integrada a simulaçôes dinámicas no contexto da Industria 4.0, focando na implementaçâo de um sistema para a análise dinámica de um processo de reaçâo química. O sistema é projetado para otimizar a análise de dados e os processos de tomada de decisao em ambientes industriáis de grande escala.