TL;DR
This paper introduces a novel spectral absorption method for real-time prediction of methane content in circulating hydrogen during desulfurization processes. This method offers high accuracy and efficiency, crucial for maintaining the purity of hydrogen, which directly impacts the quality of ultra-low sulfur diesel fuel production.
What Problem Does This Paper Address?
The petrochemical industry is increasingly focused on producing ultra-low sulfur diesel fuel to meet environmental regulations. A key challenge in this process is maintaining the purity of circulating hydrogen used in hydrotreating units. Methane, an inert component, can accumulate in this hydrogen stream, reducing its effectiveness and potentially impacting the quality of the final diesel product. Existing methods for methane detection are often slow or unsuitable for real-time industrial applications. This paper addresses the need for an accurate, efficient, and real-time method to predict methane content changes in circulating hydrogen, enabling better process control and optimization.
Methods
The core of this research is the development and application of a spectral absorption method for predicting methane content. This technique leverages the principle of Lambert-Beer's Law, where the absorption of light by methane gas is measured. Specifically, the authors utilize a monochrome light source (a 1310 µm laser) and an optical fiber coupling system to detect changes in light intensity as it passes through the circulating hydrogen. This allows for real-time measurement of methane concentration. The method involves analyzing the relationship between light intensity, gas turbidity, and path length to quantify methane. The authors also conducted experimental analysis by monitoring methane content in a real-world hydrotreating unit under various operating conditions.
Datasets
The study utilizes data collected from a hydrotreating unit in a petrochemical company in China. This real-world operational data, collected from March to November 2017, served as the basis for experimental analysis and validation of the prediction method. The paper does not introduce a new benchmark dataset but rather uses operational data to demonstrate the effectiveness of their proposed method.
Key Insights and Ideas
- Methane as a Purity Indicator: The paper clearly establishes methane content as a critical indicator of circulating hydrogen purity, directly influencing the efficiency of desulfurization and the quality of the final diesel product.
- Spectral Absorption's Advantage: The spectral absorption method, particularly with optical fiber coupling, offers a significant advantage over traditional methods by enabling real-time, non-intrusive, and safe detection of methane.
- Key Influencing Factors: The research identifies several key factors that influence methane content, including the operational time of the unit, reaction temperature, the purity of new hydrogen, feed oil density, and catalyst activity.
- Feed Oil Density Nuance: A particularly interesting finding is that within a specific range of feed oil density (892-909 kg/m³), the composition changes are minimal, and the impact on circulating hydrogen methane content is negligible. This suggests a specific operating window where this factor is less critical.
- Catalyst Activity Correlation: Higher hydrodesulfurization activity of the catalyst leads to increased methane content in the recycled hydrogen, likely due to enhanced hydrocracking reactions.
Key Results
- High Accuracy: The proposed spectral absorption method demonstrated high accuracy, with prediction errors of the methane content being less than 1.0%.
- Efficient Prediction: The method proved to be significantly more efficient than other models (Grey System Theory and BP Neural Network), with an average prediction time of 1.95 seconds compared to the others' average times of 5.47s, 4.37s, and 7.33s.
- Real-time Monitoring: The experimental results showed that the methane content in circulating hydrogen generally increased with operational time, with specific fluctuations observed due to hydrogen purging and unit maintenance.
- Factorial Influence: The study quantified the impact of various factors:
- Increased methane content in new hydrogen and higher reaction temperatures lead to increased methane in circulating hydrogen.
- Higher catalyst activity results in higher methane content in recycled hydrogen.
- Methane solubility has an inverse relationship with circulating hydrogen methane content.
Limitations and Future Work
- Laser Cost: The authors acknowledge that the adjustable laser used in their setup is expensive, although they highlight the 1310 µm laser as a more affordable and stable alternative with good application range.
- Control Strategies: While the paper identifies factors influencing methane content, the discussion on control strategies is somewhat general. Future work could delve deeper into optimizing these strategies for specific operational scenarios.
- Broader Applicability: While validated in a specific hydrotreating unit, further studies could explore the applicability of this method across different types of desulfurization processes and petrochemical plants.
- Integration with Process Control: The real-time prediction capability opens avenues for direct integration with automated process control systems to dynamically adjust operating parameters and maintain optimal hydrogen purity.
Summary updated