A. A. Ezzat, Pharos University, Alexandria, Egypt
In the petrochemical industry, business leaders are continuing to push for new approaches to optimize technology applications without disrupting their operational targets and key performance indicators (KPIs).
The massive volume of data generated and shared can be critical in plant operations. The introduction of the Industrial Internet of Things (IIoT) has presented a beneficial solution for process plants, especially petrochemicals, which are always focused on maintaining and increasing safety, reliability and efficiency, with maximum asset utilization.
The IIoT is considered a tool for partially responding to work disruptions and optimizing operational parameters. The application of modern intelligent data integration and digital twin models help operators to monitor profitability as they perform functions by balancing unit-level costs with unit-level KPIs. This has resulted in a positive approach to business sustainability—i.e., data processing can lead to risk mitigation, enhanced plant safety, increased reliability and operational efficiencies.
Moreover, to protect asset health, plant predictive analytics can perform the necessary calculations to assist in early predictions of equipment problems.
Refining-petrochemical integration optimization. With continued market demands (e.g., crude oil pricing, environmental regulations), the refining and petrochemical industries are seeking optimized solutions to ensure their survival and sustainability. Integrating the refinery and petrochemical industries enhances the synergies of existing opportunities between both industries to generate value for the entire crude oil production chain, applying economic and environmental aspects.
FIG. 1 depicts an example of an integration scheme that clarifies the relationship between refinery streams and their related petrochemical products. From data analysis, it can be concluded that due to the increased volume of data processing and the increased complexity of integrated complexes, the processing industry is facing the following challenges in plant operations and value chain optimization:
Therefore, due to the massive influx of complex data that is difficult to optimize for production control, artificial intelligence (AI) has been introduced to partially solve these challenges, enabling better process control. In addition, an optimized value chain process has led to the enhancement of the entire supply chain. This is established by leveraging historical and real-time data to generate accurate forecasts and efficient network-wide plans to attain maximized profitability.
Supply chain application. An enhanced planning approach on the value chain process has been introduced that can attain maximum profitability and KPIs. With improved visualization of historical and real-time market data, this approach can build an optimized supply chain network that can generate accurate forecasts.
This optimized supply chain networks should provide better future forecasting and planning, covering working capital reductions, production throughput and on-time orders. The expected outcomes can include:
FIG. 2 shows asset optimization solutions in a polymerization process, where target polymer production/scheduling is based on the design production slate and calculated downtimes between grade changes. This is performed by reducing transitions and addressing the real needs of polymer producers with decreased costs and new polymer market requirements. The following parameters should be considered:
IIoT solutions. With challenges in the petrochemical industry and related high data dimensions and time lags in ML applications, the introduction of the IIoT has presented a useful solution for petrochemical complexes. This solution is focused on the maintenance, safety, reliability and efficiency of processes, leading to an acceptable sustainability level. However, challenges remain, especially regarding system cybersecurity and the poor management of collected data. Therefore, to achieve a high level of business sustainability, there should be an advanced approach to utilize collected and generated plant data that can be easily translated into action. Such data should be processed and generated across cybersecurity layers. AI applications have been introduced as an optimum solution. The use of AI in the petrochemical industry has resulted in improved operational efficiency, reliability and safety. It has also provided tools for asset and value chain optimization, performance and production.
The introduction of digital twin models in the petrochemical industry has been considered, which can integrate ML tools and real-time big data. Deciding to use digital twin process applications can be divided into three primary sections: economic evaluation, investment costs and internal rate of return (IRR).
The first step for applying a digital twin system is to collect operating plant data. To ensure outcome accuracy, data validation should follow to detect any reconciliation issues or gross errors.
The second step is to perform dynamic target calculations considering feedrate, quality, composition and independent parameters, while also meeting selected product rates, specifications and environmental regulations.
The third step is to embed the optimized system into the entire supply chain of the production complex. The application of the digital twin system should solve the following:
For example, inputting petrochemical data (such as a polymerization reactor’s control, along with feedstocks, catalyst used, cooling medium and reaction time) into a digital twin system can help operators understand how these factors are working in relation to impacts on the process’ profitability and catalyst performance in a safe and optimized manner (FIG. 3). These advanced digital application solutions integrate real-time data and advanced analytics for better decision-making, underpinning applications that can dramatically improve process control behavior, efficiency and sustainability.
With digital twin systems, plants can now achieve a new level of business sustainability—one that improves not just environmental impacts, but also asset performance, regulatory compliance and financial performance. These systems will also help organizations with business transactions, allowing operators to test economical and efficient processing outcomes in their operations.
From the above material, it can be visualized that digital twin system applications in the petrochemical industry are considered a system integration approach through engineering, procurement, construction management and other key services related to supply chain and marketing activities. These applications are based on a developed operating control system, driving plant sustainability operations to attain the necessary visibility analysis and insights to address the challenges inherent to meeting process goals.
Takeaway. Traditionally, industrial information technology (IT) and operational technology (OT) have been considered separate entities with little crossover. However, hardware and software systems that monitor plant operations and optimization studies provide operational control that has been engineered and supported by data (IT).
In today’s digital world, the interconnection between IT and OT is providing the processing industry with valuable data to attain high KPIs. This is based on the IIoT, ML and loop design for information exchange between the factory and a virtual digital model.
End users can now leverage existing assets and investments to drive toward a safer, more reliable and efficient petrochemical enterprise. Therefore, the application of a digital twin model can result in a more secure, safer and more efficient petrochemical complex. HP
ABBAS EZZAT works as a Petrochemical Professor at Pharos University and a Distinguished Scientist at the Materials Science Department within the Institute of Graduate Studies and Research at Alexandria University. He is also a local consultant for the Egyptian petroleum and petrochemical sectors and a Senior Associate Consultant for Channoil Consulting Ltd. in London. Prior to joining academia, he occupied several top management positions in the Egyptian petroleum and petrochemical industries. Dr. Ezzat holds an MSc degree in chemical engineering from Washington University and a PhD in petrochemical applications from Alexandria University. He completed his postgraduate studies in petroleum processing technologies from the School of Chemical Engineering at Oklahoma State University.