May-2024

Optimising furnace run length in a steam cracker using AI

Case study on improving ethylene furnace run length by leveraging the synergy of digitalisation and artificial intelligence to provide the necessary insights.

Surabhi Thorat and Vivek Srinivasan, Dorf Ketal Chemical (I) Pvt Ltd
Sudarshan Vijayaraghavan, Dorf Ketal Chemicals PTE Ltd

Article Summary

Ethylene serves as a predominantly petrochemically derived monomer essential for producing plastics, fibres, and various organic chemicals. These end -products find applications across industries such as packaging, transportation, construction, and other industrial and consumer markets. Notably, more than half of global ethylene derivative consumption is attributed to non-durable or consumable end uses, especially in packaging.

The majority of this consumption is associated with polyethylene, a plastic resin that constitutes most ethylene usage. Given its status as one of the largest volume petrochemicals globally, the consumption of ethylene is influenced by economic and energy cycles. Its extensive and diverse derivative portfolio, covering both non-durable and durable end uses, positions ethylene as a benchmark for gauging the overall performance of the petrochemical industry.

Furnace efficiency
In the complex realm of ethylene production, furnace efficiency plays a pivotal role in determining overall operational success. However, a significant challenge arises from shortened run lengths, increasing downtime and production costs. This issue is rooted in the increasing tube metal temperature (TMT) of the furnace, a critical factor affecting operational longevity.

A decrease in run length in an ethylene furnace due to coke formation and a rise in TMT can result in significant financial losses for several reasons. Here are some potential factors contributing to the financial impact:
• Production interruption: Reduced run length means the ethylene furnace is not operating at its optimal capacity for the intended duration. This interruption in production can lead to lower yields of ethylene and other desired products, resulting in lost revenue.
• Increased decoking: Coke formation and elevated TMT may call for frequent decoking, resulting in higher energy costs and reduced capacity utilisation rates. Frequent decoking can also affect the overall reliability of the tubes, compromising the life span of radiant tubes.
• Energy consumption: A less efficient furnace may require more energy to maintain the desired operating conditions. Higher energy consumption not only leads to increased operational costs but also contributes to environmental concerns if the energy source is not sustainable.
• Impact on selectivity: Inefficient furnace operation, combined with a high potential for coking, leads to a critical TMT threshold during mid-run cycles, constraining achievable severity levels. Consequently, this limitation adversely affects the overall selectivity for ethylene make.
• Market dynamics: In the competitive petrochemical industry, delays or interruptions in production can affect a company’s ability to meet customer demands. This can result in lost market share and potential long-term damage to business relationships.
• Increased emission and environmental impact: Poorly optimised furnace operations tend to emit higher levels of greenhouse gases, thereby exacerbating the carbon footprint, as furnaces are commonly fuelled by fossil fuels. The frequent decoking process also introduces further emissions, compounding an already elevated environmental impact.

To mitigate these financial losses, it is crucial for plant operators to implement effective monitoring, maintenance, and operational strategies to prevent or address issues such as coke formation and elevated TMTs in a timely manner. Regular inspections, proper decoking procedures, and adherence to best practices in furnace operation can contribute to improved efficiency and extended run lengths.

Differentiator
Understanding and addressing elevated TMTs is key to overcoming these aforementioned challenges. Dorf Ketal’s proprietary artificial intelligence (AI) solution CokeNil offers solutions to extend run lengths through predictive maintenance, optimised process control, and improved fault detection. It is a data-driven, deep domain insight-based solution where every process parameter is evaluated thoroughly in the exploratory data analysis phase to ensure the accuracy of the outcome. It is built on advanced deep learning methods such as long short-term memory (LSTM) networks, Random Forest regression, CatBoost regression, XGBoost regression, and time series algorithms to help the model identify the pattern and behaviour of each critical process parameter.

CokeNil is a unique AI solution for optimising the furnace run length. This plant’s distributed control system (DCS) feeds live operating conditions (such as naphtha feed composition, coil outlet temperature, steam-to-hydrocarbon ratio [SHC], temperature, and fuel flow) to the CokeNil. These are processed by the AI model, and optimal values that need to be set in the real process are recommended. A key point is that the optimal values will never violate the acceptable operational ranges of suggested critical parameters.

AI algorithms equipped with real-time data analysis capabilities play a crucial role in predicting and proactively addressing potential furnace issues. By dynamically adjusting parameters and ensuring adaptive control, the AI system prevents runway TMT and mitigates stress on critical components. This dynamic optimisation lays the groundwork for extended operational runs, minimising the interruptions caused by unplanned downtime.

AI/digital twins symbiosis
AI-driven optimisation plays a key role in ensuring stable furnace operation with reduced delta pressure and dilution steam, as well as an optimised feed ratio. This leads to a more controlled furnace temperature, contributing to operational steadiness. The nuanced control provided by AI not only enhances operational efficiency but also contributes significantly to the overall stability of the ethylene production process.

The discussion takes an intriguing turn with the introduction of advanced process modelling, including the concept of digital twins. This innovative approach enables virtual testing and scenario optimisation, allowing operators to explore various conditions without impacting the physical furnace. The symbiosis of AI and digital twins not only optimises current operational parameters but also lays the groundwork for future advancements in ethylene production processes.

However, challenges persist, and historical data serves as both a hurdle and a guiding light. The initial phase of AI model building includes critical process parameters identification, preliminary investigations, pattern discovery, anomaly spotting, hypothesis testing, and establishing correlations between process parameters and run length by considering TMT as the primary target.

AI leverages historical data to train the run length optimisation model. Coking in the radiation coil is inevitable at high cracking temperatures. As TMT reaches maximum threshold values, the furnace undergoes decoking, compromising run length. Furnace run length is directly linked to coke formation rate, with operational parameters such as TMT, dilution steam ratio, firing rate, feed composition, feed rate, wall and floor burners in operation, excess oxygen %, coil pressure ratio (CPR), coil outlet pressure and venturi ratio determining the end of furnace runs.

In response to these intricacies, the furnace is operated, and parameters are controlled to minimise the rate of increase in TMT throughout the run. This proactive approach mitigates the rise in TMT, ultimately contributing to longer furnace run lengths and improved operational efficiency.