Digitalization can help ethylene producers to flexibly respond to economic uncertainties and assess available options based on their existing infrastructure and priorities
Post-pandemic, olefin and polyolefin producers have been faced with regional and global economic headwinds. Inflation, energy security, rising cost of energy and feedstock, regional conflicts and destocking have impacted their market. To respond to market uncertainties, the industry has worked effectively to close margin gaps through improving operational performance, lowering energy use, and increasing agility to respond to fluctuations requiring rate cuts.
On the environmental front, many ethylene producers have committed to curb emissions by 2030 and 2050 milestones. Such avenues for decarbonization include large-scale electrically heated steam cracker furnaces, such as the one currently under construction at BASF’s Ludwigshafen Verbund site in Germany.
Circularity via the mechanical and chemical recycling of waste plastics is another priority for the industry. Renewable plastics (derived from bio-based feedstock or recycled pyrolysis gas) have a lower carbon footprint than ethylene produced in the traditional steam-cracking process. Even though in the short term, the demand for virgin feedstock may not change, in the long-term, product circularity will lower the need for traditional fossil-based feedstock.
Digitalization has become integral in ethylene producers’ business strategy. Digitalization improves plant performance, efficiency, reliability, safety and product margins, while also playing a key role in developing new technologies for sustainability initiatives. Designing new processes for chemical recycling of plastics, incorporating bio-based and recycled feedstocks, and integrating renewable energy sources are just a few examples. Digital technologies are also necessary for operator training to improve plant safety and expedite planned shutdowns and startups (Figure 1). Leveraging digital solutions, new sustainability project trade-offs can be evaluated, and risks minimized. As the ethylene production landscape is evolving to meet the growing demand, digitalization helps producers balance growth and sustainability.
FIGURE 1. Manufacturing sites are increasingly turning to digital technologies to aid in their training regimens around energy management, plant safety and startups and shutdowns
Digital solution framework
When it comes to production digitalization, there is a broad spectrum of applications that can improve key performance indicators (KPIs). However, this vast range of solutions can be categorized into three main groups — those contributing to energy efficiency and emissions reduction; those contributing to improving production performance; and those accelerating overall sustainability. In the following sections, only one or two use cases per segment are discussed. Table 1 provides a framework for the applicability and impact of some of the tools and use cases.
Energy efficiency and emissions
There are many digital tools available to help plants optimize energy efficiency and reduce emissions production. A few categories are described in the following sections.
Digital planning and scheduling tools. Ethylene producers use digital planning and scheduling (P&S) tools to determine optimal combinations of feedstock, cracking severity and furnace lineups. P&S tools enable them to achieve higher margins while lowering energy use and responding to market opportunities. For multi-site operators, choosing the correct combination of feedstock at each site, production rates and determining marginal ethylene production allocations can maximize gains. These plans are robust and flexible and have lower execution risks. Modern P&S tools enabled with artificial intelligence (AI) and machine learning (ML) capabilities can be trained based on historical data to accelerate schedule creation and initialize optimization. AI-powered interactive advisors can also pro-actively advise planners based on operational constraints.
P&S tools can be integrated with advanced process control and dynamic optimization to provide unique capabilities to operate plants profitably while enabling quick responses to changes in feedstock and product markets. As an example, when an ethylene producer located in the U.S. Gulf Coast region had to cut rates due to an impending hurricane, they minimized lost opportunity by running the optimal feedstock during the rate-cut period. Another producer leveraged production planning optimization to select the optimal feedstock when faced with reliability problems on one of their furnaces, therefore minimizing lost opportunity.
With growing environmental priorities to mitigate emissions and improve energy efficiency, planning and scheduling tools play an additional role in reducing carbon footprints. In a recent study, a major Middle Eastern ethylene producer has used a multi-plant planning model to explore the operational levers available at plant sites to reduce the CO2 footprint of its network of crackers. The study showed that it is possible to decrease cracker CO2 emissions significantly (around 3–5%) by using common operational degrees of freedom without incurring capital expenditures and with minimal impact on olefins production and operating margin. The study explored specific plant economics, operational constraints and flexibility of each plant to achieve an optimum CO2 emissions reduction.
Advanced process control and dynamic optimization tools. Advanced process control (APC) tools have been essential parts of operations for decades. In difficult markets, asset owners direct their investments from larger capital investments and new builds to optimizing and improving productivity, efficiency, sustainability and safety of their existing facilities. APC is one of the strongest tools to achieve economic, safety and sustainability goals with minimal upfront investments. Modern APC tools have expanded capabilities to fit into a wider range of operating conditions. They are self-tuning (adaptive) and self-calibrating thanks to deep learning and AI-embedded functionalities, easier to use and cost less to implement and maintain. While economic objective functions of APC enable producers to run operations to the best of their interest, less experienced engineers can benefit from less complex and AI-guided APC features to run the operation.
In an ethylene plant, APC can be used for a variety of functions. Closing the gap between planning/scheduling and operation, increased agility, following aggressive plans, reducing margin leakage, reducing flaring and CO2 emissions and finally setting the stage for greater autonomy are some of these benefits. For example, when it comes to energy efficiency, a modern APC can model process parameters to execute energy-efficient control actions, update models online with non-disruptive background testing and stabilize processes by minimizing impact of disturbances and process fluctuations on energy use.
Reducing cracking furnaces’ fuel use and emissions is an excellent APC use case. Cracking furnaces are the most energy-intensive components of an ethylene plant. A reduction in furnace fuel use by a few percentage points translates to millions of dollars in annual savings, while cutting back on large amounts of CO2 emissions. In a 2019 study, A major Latin American ethylene producer revamped their APC controller in one of their cracking furnaces based on a new control strategy leveraging adaptive control tools. They were able to significantly curb fuel usage (Figure 2), leading to $1.15 million in savings per year. Later, the model was expanded to another five crackers.
FIGURE 2. Cracker fuel gas usage is compared before and after the new APC implementation. The crackers on average saved 5% fuel gas with the new APC
Cracked gas compressors (CGCs) can also benefit from APC. An estimate of this benefit is presented in Table 2, based on the reduction in suction pressure. A reduction of 0.1 kg/cm 2 in suction pressure results in 0.1% increase in ethylene yield, equivalent to about 2 ton/d more of ethylene production.
An olefin plant optimization happens in multiple layers and at different time scales. The planning and scheduling layer, as mentioned earlier, focuses on optimizing the feedstock selection, inventory and furnace swaps monthly or weekly. On the other hand, the APC layer focuses on optimizing feed rate and energy, and meeting product quality on a minute-by-minute basis. In between the planning and APC layers is the middle layer of optimization. It typically involves optimizing feed allocation to the different furnaces, severity/conversion optimization, steam-ratio optimization and suction-pressure optimization performed at the same frequency as the APC layer. Integrating all three layers creates a “unified” platform which streamlines multi-user workflows. While running the plant to its limits, a unified platform can reduce margin leakage by tracking plan execution in real-time. Figure 3 represents KPIs and optimization layers in olefins production.
FIGURE 3. KPIs and optimization layers for the operation are shown. APC and dynamic optimization layers respond to minute-by-minute changes, while planning and scheduling (P&S) has a longer-term outlook
Multivariate analysis. Multivariate analytical tools for optimization involve data-driven models that provide quick, actionable insights from historical data. Multivariate analysis tools can combine historical and planning data to identify variables with the greatest impact on deviations (for example, energy consumption or product quality). Identifying these variables enables plant engineers and operators to take the necessary actions to adjust operations.
One use case for multivariate analysis is analyzing cracking furnace energy use. In one study, the influence of gas distribution between the side wall and the bottom burners of the furnace was evaluated. Similarly, the best relationship between feed rate and fuel gas consumption was mapped. A range of factors, such as ambient conditions, sensor conditions, feedstock and the specific products can potentially result in operating levels outside of the optimal ranges. In the above example, around 50 to 70 kg/h of fuel gas were saved after implementing the study conclusions.
Improving steam-turbine efficiency is another use case for multi-variate analysis. CGCs are large consumers of energy and improving their efficiency lowers steam use and energy cost. However, turbine operation is typically complicated and depends on many factors, such as gas composition, ambient conditions, upstream and downstream operations, performance curves and control targets. With constantly changing demand patterns and non-linear relationships, traditional techniques for calculating efficiency seldom reflect the real world. In a recent study, a global petrochemical company used multivariate analysis to create a single model around turbine and CGC operation, capturing dynamic process behavior and identified factors contributing to efficiency loss. By adjusting these factors, the producer managed to save 5% on turbine steam consumption.
Production performance
Digital twins and real-time optimization are two tools that engineers can turn to when looking to improve production performance. Digital twins are widely used in industry to provide a digital representation of physical assets and their operations. If the use case is general evaluation and reporting, then an offline steady-state model can be sufficient. However, digital twins can become extremely powerful tools when they receive process measurements directly from the distributed control system (DCS) or plant historian and use the data to optimize the plant and make recommendations for change in real-time (open loop).
Online digital twins are extremely useful in olefin plants due to the complexities of the process units and processed feedstocks. Changing feed slate, product profitability and pricing, considerations of energy use, and furnace swaps, among other factors, make real-time monitoring and optimization a necessity. Online optimization can increase the yields of more valuable products, reduce energy consumption per ton of feed or product, push a plant closer to multiple true constraints and increase throughput while evaluating non-linear trade-offs like furnace yields versus refrigeration or distillation.
A major ethylene producer located in the U.S. Gulf Coast region has been using online digital twins for more than 15 years to make trade-offs at the unit level and to maximize gross margins. The site has been able to leverage the tool to improve yield. For instance, when a compressor discharge flow did not match the online model, it led to the discovery of a minimal flow leakage. Fixing the leakage increased the production rate by 1.1%.
Accelerating sustainability
Many ethylene producers have set sustainability milestones along their decarbonization journey for the coming decades through 2050. Improving energy efficiency and lowering emissions through optimizing operations perhaps has the highest and most immediate impact. However, to truly reshape the industry, many producers are evaluating a diverse range of new technologies.
Leveraging bio-based or recycled feedstocks to produce ethylene is perhaps the next logical step to decarbonize ethylene production. It is estimated that bio-based ethylene production (derived from bio-ethanol) can lower up to 40% of greenhouse-gas emissions associated with ethylene production [1]. Chemical upcycling of plastic waste or using sustainable lignocellulosic biomass to produce bio-ethylene are two effective and challenging pathways to reduce production emissions. Digital solutions are fundamental to design, de-risk, scale and understand the technical and economic trade-offs of these processes. Additionally, in markets requiring circular and low-carbon-footprint products, an accurate mass balance of the attributed renewable feedstock is essential. This accurate balance enables producers to obtain certifications, such as REDcert2 or ISCC Plus, for the renewable feedstock used. Digital accounting and reconciliation tools here enable accurate calculations and reporting.
Based on an International Energy Agency estimate, solar, wind and energy efficiency could deliver around half of emissions reduction to 2030 [ 2]. This estimate highlights the power of electrification in energy intensive assets. However, incorporating distributed and intermittent energy sources (wind, solar, geothermal, hydro) requires an advanced distributed-energy resource-management system (DERMS). Digital-grid management solutions ensure sustainability, reliability and resiliency amidst increasingly dynamic supply and demand situations for ethylene producers.
Carbon capture and utilization (CCU) is another way that leading producers can pursue lower carbon emissions. Designing a new steam cracker with built-in capture facilities or retrofitting an existing ethylene plant to add a CCU extension are only possible through digital engineering and economic evaluation tools. Digital solutions help producers with making strategic or detailed decisions like identifying a CO2 destination or determining the required utilities. Technology licensors are leveraging digital engineering to develop CO2-capture facilities to make methanol or green ethylene to further curb emissions.
The global landscape of ethylene production is changing. Flexibility in production and agility to respond to economic uncertainties while responding to consumers’ growing needs require new levels of operational excellence. In addition, decarbonization of the chemical industry requires new technologies that don’t currently exist at the scales required. Digital solutions provide unique opportunities for the chemical industry to accomplish productivity and sustainability goals for the coming decades. ■
References
1. International Renewable Energy Agency (IRENA), Production of Bioethylene: Technology brief, January 2013.
2. International Energy Agency (IEA), Net Zero by 2050, 4th Rev., October 2021.
Author
Ana Khanlari is the solution marketing director for chemicals at Aspen Technology, Inc. (Email: [email protected]). In her role, she works with chemical manufacturers to increase value delivery and sustainability of operations, leveraging digital solutions. She has broad experience in innovation, product development, manufacturing and field applications in global chemicals markets, including olefins, specialty monomers and water treatment, both as a researcher and an industry technical consultant. Khanlari has several published articles and patents in the areas of digitalization, olefin process treatment and specialty polymers. Khanlari earned a Ph.D. in chemical engineering at the University of Kansas.