Method for precisely converting coal tar into high-cetane-number alkane through hydrogenation

Abstract

The invention provides a method for precisely converting coal tar into high-cetane-number alkane through hydrogenation, which comprises the following steps: obtaining reaction data in a preliminary process condition range, if the hydrogenation depth exceeds a preset threshold value, adjusting a temperature control parameter to reduce the decyclization degree, and obtaining stable n-alkane generation efficiency through iterative calculation; according to the adjusted reaction path scheme, a hydrogenation process is operated in an actual reactor, a cetane number fluctuation range is extracted from real-time monitoring data, and whether process conditions and catalyst gradation adaptability meet target requirements or not is judged; if the fluctuation range of the cetane number exceeds a preset threshold value, analyzing the interaction influence of temperature control and pressure regulation through regression, and optimizing process conditions by adopting a gradient descent algorithm to obtain a final process parameter combination; long-term operation data of an accurate conversion result is obtained, the stability of catalyst grading and process conditions is analyzed through a machine learning algorithm, the dynamic adjustment requirement of a reaction path is judged, and a self-adaptive hydrogenation optimization model is obtained.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method for accurately converting coal tar into high-cetane number alkanes by hydrogenation. Background Art

[0002] Background: Coal tar hydrogenation technology, as a crucial component of the coal chemical industry, plays a crucial role in promoting energy transition and meeting demand for high-quality fuel oil. With the increasing stringency of national fuel oil quality standards, particularly cetane number requirements, this technology has become a key component in achieving efficient, environmentally friendly, and high-quality production. However, widespread application of coal tar hydrogenation technology currently faces numerous constraints, and breakthroughs are urgently needed to meet the needs of industry development. Existing coal tar hydrogenation methods have exposed a series of limitations in practice. Traditional processes and catalyst formulations struggle to effectively address the complexity of coal tar composition, resulting in over-hydrogenation of polycyclic aromatic hydrocarbons (PAHs), reduced alkane content, and difficulty consistently achieving the product cetane number standard. Furthermore, extensive process adjustments and inadequate catalyst performance lead to a lack of precision in controlling the depth of hydrogenation, limiting further improvements in product quality. These issues are particularly prominent when meeting stringent national standards and have become bottlenecks in technological advancement. Specifically, the core challenges facing coal tar hydrogenation lie in the poor compatibility of catalyst grading schemes with process conditions and the difficulty in precisely controlling the hydrogenation reaction pathway. Due to the complex distribution of polycyclic aromatic hydrocarbons (PAHs), monocyclic aromatic hydrocarbons (MAHs), and alkanes in coal tar, the inadequate selective conversion of existing catalysts makes it difficult to balance ring opening and chain scission during hydrogenation, severely impacting the efficiency of normal-alkanes production. Furthermore, the lack of systematic optimization of process parameters such as temperature and pressure makes it difficult to precisely control the degree of PAH decyclization within the reactor, further exacerbating the technical challenge of unstable cetane numbers. Therefore, optimizing catalyst grading and process conditions to achieve the precise conversion of PAHs to high-cetane-number alkanes during coal tar hydrogenation has become a key issue in improving product quality and meeting market demand. Summary of the Invention

[0003] The present invention provides a method for accurately converting coal tar into high cetane number alkanes by hydrogenation, which mainly comprises: Obtain chemical composition data of coal tar samples, analyze the distribution ratios of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons, and alkanes, determine the characteristic parameters of the initial reactants, and obtain benchmark values ​​for the proportion of polycyclic aromatic hydrocarbons and the target conversion pathway; By optimizing the catalyst grading scheme, combined with temperature control and pressure regulation parameters, a process condition simulation system was constructed to calculate the dynamic changes of hydrogenation depth and ring-opening equilibrium in the virtual reactor and determine the preliminary process condition range; Acquire reaction data within the initial process conditions. If the hydrogenation depth exceeds a preset threshold, adjust the temperature control parameters to reduce the degree of decyclization. A stable n-alkane production efficiency is obtained through iterative calculation. Focusing on the stable n-alkanes production efficiency, the changing trend of the decyclization degree in the reaction pathway is analyzed. By optimizing the pressure adjustment parameters, it is determined whether the selectivity of the conversion of polycyclic aromatic hydrocarbons to n-alkanes meets the expectations, and the adjusted reaction pathway scheme is obtained. Based on the adjusted reaction pathway, the hydrogenation process was run in an actual reactor. The cetane number fluctuation range was extracted from the real-time monitoring data to determine whether the process conditions and catalyst grading were compatible with the target requirements. If the cetane number fluctuation range exceeds the preset threshold, the interactive effects of temperature control and pressure regulation are analyzed through regression, and the gradient descent algorithm is used to optimize the process conditions to obtain the final process parameter combination; Using the final process parameter combination and catalyst grading scheme, the full hydrogenation process was run, and the n-alkane content and cetane number were extracted from the product analysis data to determine the precise conversion of polycyclic aromatic hydrocarbons to high-cetane number alkanes. Obtain long-term operating data for accurate conversion results, analyze the stability of catalyst grading and process conditions through machine learning algorithms, determine the dynamic adjustment needs of the reaction path, and obtain an adaptive hydrogenation optimization model.

[0004] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses a method for accurately converting coal tar into high-cetane number alkanes by hydrogenation. The method analyzes the chemical composition of the coal tar sample to determine the proportion of polycyclic aromatic hydrocarbons and the conversion path baseline value, and combines the optimized catalyst grading scheme and the process condition simulation system to achieve dynamic regulation of hydrogenation depth and ring-opening balance. The present invention ensures the stability of the normal alkane generation efficiency by iteratively optimizing the temperature and pressure parameters, and judges the adaptability of the process conditions and catalyst grading by real-time monitoring of the cetane number fluctuation range. The process conditions are optimized using a gradient descent algorithm, and ultimately an accurate conversion scheme of polycyclic aromatic hydrocarbons to high-cetane number alkanes is obtained. The present invention also establishes an adaptive hydrogenation optimization model to achieve dynamic adjustment of the reaction path. The method effectively solves the problems of insufficient catalyst selectivity and lack of process parameter optimization in the coal tar hydrogenation process, significantly improves the stability of the product cetane number and the normal alkane content, and provides key technical support for the production of high-quality fuel oil in the coal chemical industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 The present invention provides a flow chart of a method for accurately converting coal tar into high-cetane number alkanes by hydrogenation.

[0006] Figure 2Schematic diagram of a method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to the present invention.

[0007] Figure 3 This is another schematic diagram of the method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to the present invention. DETAILED DESCRIPTION

[0008] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0009] like Figure 1-3 In this embodiment, a method for accurately converting coal tar into high cetane number alkanes by hydrogenation may specifically include: S101. Obtain the chemical composition data of the coal tar sample, determine the characteristic parameters of the initial reactants by analyzing the distribution ratios of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons, and alkanes, and obtain the baseline values ​​of the polycyclic aromatic hydrocarbon content ratio and the target conversion path.

[0010] Coal tar samples are tested using spectral analysis to obtain raw distribution data for polycyclic aromatic hydrocarbons (PAHs), monocyclic aromatic hydrocarbons (MAHs), and alkanes. Based on this raw distribution data, the distribution ratios of each component are calculated, and a pre-set classification model is used to determine their relative content. If the PAH content exceeds a preset threshold, information processing is used to prioritize the conversion pathways and generate a preliminary plan. Based on this preliminary plan, the alkane distribution data is combined to determine the path adjustment direction and determine the final target conversion pathway baseline value.

[0011] For example, when testing coal tar samples using spectral analysis techniques, ultraviolet-visible spectroscopy or infrared spectroscopy can be used to obtain the raw distribution data of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons, and alkanes in the sample. The principle of spectral analysis is to detect the sample's absorption or reflection characteristics of light of specific wavelengths, identify the characteristic peaks of different chemical bonds or functional groups, and thus infer the type and approximate content of each component. Suppose that in a test, polycyclic aromatic hydrocarbons show a strong absorption peak in a specific band, indicating that they may account for a high proportion, while the characteristic peaks of monocyclic aromatic hydrocarbons and alkanes are relatively weak.

[0012] In a possible implementation, when calculating the distribution ratio of each component based on the original distribution data, it can be estimated by integrating the spectral peak area.

[0013] For example, the peak area of ​​PAHs accounts for 50% of the total area, monocyclic aromatics 30%, and alkanes 20%. Subsequently, using a pre-defined classification model, such as a machine learning-based content prediction model, these proportion data are input into the model to determine the relative content of each component. Assume that the model outputs 55% PAHs, 25% monocyclic aromatics, and 20% alkanes, which is consistent with the original data trend.

[0014] For example, if the preset PAH threshold is 40%, and the test results show a PAH content of 55%, significantly exceeding the threshold, the information processing phase will be required to prioritize the conversion pathways and form a preliminary plan. The principle of information processing is to use historical data and process requirements to prioritize conversion pathways that can effectively reduce PAH content, such as hydrocracking, which can convert polycyclic structures into monocyclic or alkane structures. The preliminary plan may prioritize the hydrocracking pathway and combine it with suggestions for optimizing reaction conditions.

[0015] In one possible implementation, based on the preliminary plan, when combining alkane distribution data to determine the direction of path adjustment, if the alkane proportion is only 20%, it indicates that the alkane yield may need to be further increased during the conversion process to balance the product distribution. Therefore, the adjustment direction may tend to increase the cracking depth or optimize the catalyst selection, and ultimately determine the target conversion path baseline value. For example, the reaction temperature of the hydrocracking process is set within a specific range to ensure the expected PAH conversion rate while increasing the alkane proportion to above 30%. The beneficial effect of this is that it not only reduces the environmental risks of PAHs, but also increases the proportion of high-value alkane components in the product.

[0016] For example, from another perspective, if the alkane distribution data shows that its proportion is close to the ideal value, the adjustment direction may focus more on the precise conversion of polycyclic aromatic hydrocarbons to avoid excessive cracking and energy waste. The final path benchmark value may focus on mild reaction conditions to ensure conversion efficiency while reducing energy consumption. This multi-angle analysis ensures the flexibility and adaptability of the solution and significantly improves the economy and environmental friendliness of the process. Through the above method, not only can the distribution of each component in coal tar be accurately controlled, but it can also provide a reliable basis for subsequent industrial applications, reflecting the dual value of technology in resource utilization and environmental protection.

[0017] S102. By optimizing the catalyst grading scheme and combining the temperature control and pressure adjustment parameters, a process condition simulation system is constructed to calculate the dynamic changes of hydrogenation depth and open-ring equilibrium in the virtual reactor and determine the preliminary process condition range.

[0018] By constructing a simulation system, reaction data for the catalyst grading scheme under different temperature and pressure parameters is obtained to determine preliminary dynamic change trends. Based on these dynamic change trends, a virtual reactor is used to simulate the hydrogenation depth and ring-opening equilibrium in real time, identifying the critical fluctuation range during the reaction process. Within this critical fluctuation range, a control combination of the temperature and pressure parameters is obtained. Using a pre-established regression model, the impact of this control combination on the reaction simulation is analyzed to determine the optimal control range and optimize the reaction process.

[0019] For example, when building a simulation system to obtain reaction data for a catalyst grading scheme, one can simulate the reaction environment under different temperature and pressure conditions to gain a preliminary understanding of the catalyst's performance in the coal tar hydrogenation process. For the temperature parameter, a range of, for example, 300 to 450 degrees Celsius and the pressure parameter between 5 and 15 MPa can be set. The simulation system then records the changes in catalyst activity and product distribution under these conditions. This approach can help identify the effects of temperature and pressure on reaction rates, providing data support for subsequent dynamic trend analysis.

[0020] For example, when analyzing dynamic trends, we can focus on two key indicators: hydrogenation depth and ring-opening equilibrium. Suppose that under a certain temperature and pressure combination, the hydrogenation depth shows a gradual increase, while the ring-opening equilibrium fluctuates significantly within a certain range. Through real-time simulation of the virtual reactor, we can observe that the hydrogenation depth accelerates as the temperature rises, while the ring-opening equilibrium may stabilize as the pressure increases. This simulation helps identify the key fluctuation ranges in the reaction process and provides a basis for subsequent regulation.

[0021] For example, analyzing control combinations within critical fluctuation ranges allows for optimizing reaction conditions by adjusting temperature and pressure parameters. For example, at a temperature of 400°C and a pressure of 10 MPa, the fluctuation range is relatively small, while the hydrogenation depth reaches a high level. A pre-established regression model can be used to analyze the impact of this combination on the reaction simulation. The model might reveal that for every 10°C increase in temperature, the hydrogenation depth increases by one percentage point, while changes in pressure have a more significant impact on the ring-opening equilibrium. This analysis can help determine the optimal control range, ensuring a smoother and more efficient reaction process.

[0022] Specifically, when optimizing a reaction process, the feasibility of a control combination can be verified from multiple perspectives. For example, by comparing product distributions under different temperature and pressure combinations, the conditions for the most efficient conversion of polycyclic aromatic hydrocarbons to monocyclic aromatic hydrocarbons can be determined. Alternatively, by simulating the catalyst's deactivation rate, it can be determined whether the catalyst's service life within a certain control range can meet production requirements. These multi-faceted analyses support each other and collectively point to a more optimal combination of reaction conditions, helping to improve overall process stability.

[0023] It should be noted that real-time simulation of the virtual reactor plays an important role in this process. It not only dynamically reflects the impact of changing reaction conditions on hydrogenation depth and ring-opening equilibrium, but also provides intuitive data reference for subsequent regulation.

[0024] For example, in a simulation scenario, when the temperature approaches the upper limit, an increase in byproduct formation may be observed, which can be effectively suppressed by reducing the pressure. This feedback mechanism provides a reliable basis for process optimization while also reducing the risk and cost of actual experiments.

[0025] It’s important to note that the application of regression models makes the analysis of the impact of control combinations more systematic. By combining historical and simulation data, the model can predict reaction performance under different parameter combinations, providing scientific guidance for process adjustments.

[0026] For example, within a specific control range, the model may suggest the optimal ratio of temperature and pressure, thereby minimizing energy consumption while maintaining conversion efficiency. This approach has important practical value in the field of coal tar processing, significantly improving the economics and operability of the process.

[0027] S103. Acquire reaction data within the initial process condition range. If the hydrogenation depth exceeds a preset threshold, adjust the temperature control parameters to reduce the degree of decyclization, and obtain a stable normal alkane production efficiency through iterative calculation.

[0028] Initial reaction data is obtained from the process conditions and standardized using data processing tools to generate a preliminary, organized reaction data set. Based on this data set, real-time monitoring of the hydrogenation depth is implemented. If the hydrogenation depth exceeds a preset threshold, a condition adjustment mechanism is triggered to determine the adjustment range of the temperature parameter. Within this adjusted temperature parameter range, real-time data on the degree of decyclization is obtained. A regression analysis model is used to fit this data to determine whether the degree of decyclization has stabilized.

[0029] For example, during the acquisition of the initial reaction data of the process conditions, various indicator data in the reactor, such as temperature, pressure, and raw material flow rate, can be collected in real time through sensors. These data often have certain deviations due to different collection equipment or environmental interference, so they need to be standardized. Standardization can be understood as the process of converting data of different dimensions into a unified standard, such as unifying temperature data from degrees Celsius to a dimensionless numerical range for subsequent analysis. Taking a specific scenario as an example, assuming that the temperature range in the initial reaction data is between 200 and 300 degrees Celsius, it may be mapped to the range of 0 to 1 after standardization, which is convenient for unified comparison with other parameters such as pressure. This processing method helps to reduce the impact of data noise on subsequent analysis.

[0030] For example, in real-time monitoring of hydrogenation depth, a preset threshold can be set to determine whether the reaction is within the expected range. Assuming the ideal hydrogenation depth range is 80% to 90%, if the monitoring data shows that the current hydrogenation depth is only 75%, it indicates that the reaction may not be achieving the expected effect. In this case, the trigger condition adjustment mechanism becomes particularly important.

[0031] Specifically, historical data analysis can be used to determine the adjustment range for temperature parameters, such as increasing the temperature from 250°C to 260°C to see if this effectively increases the depth of hydrogenation. This real-time monitoring and dynamic adjustment approach allows for timely identification of problems and the implementation of measures to ensure the stability of the reaction process.

[0032] For example, when acquiring real-time data on the degree of decyclization and fitting a regression analysis model, a trend model can be constructed using data from multiple experiments. Assuming that after temperature adjustment, the decyclization data shows a trend of gradually increasing from an initial 60% to 75%, regression analysis can be used to fit a curve to determine whether the degree of decyclization has stabilized. If the data fluctuates slightly between 72% and 75%, it can be considered to be close to a stable state. This analysis method can intuitively reflect the impact of process parameter adjustments on reaction results, providing data support for subsequent optimization. Furthermore, this fitting analysis can help predict reaction trends over a period of time, providing a reference for further adjustments to process conditions.

[0033] For example, judging the stability of the decyclization degree can be done through comparative analysis based on historical cases. For example, if, in previous experiments, the reaction efficiency reached a high level when the decyclization degree stabilized at 74%, then if the current data approaches this value, it can be preliminarily determined that the process conditions are close to ideal. This comparative analysis can provide empirical support for fine-tuning process parameters while reducing unnecessary testing costs.

[0034] For example, throughout the entire process of process optimization, the standardization of initial data, monitoring of hydrogenation depth, and analysis of changes in the degree of decycling form a complete closed-loop system. The implementation of each step provides a reliable data foundation for subsequent steps. Furthermore, through dynamic adjustments and model analysis, the control precision of the reaction process can be effectively improved. This multi-step collaborative approach not only improves data processing efficiency but also lays the foundation for continuous improvement of process conditions.

[0035] S104. Analyze the changing trend of the degree of decyclization in the reaction pathway with respect to the stable n-alkanes production efficiency. By optimizing the pressure adjustment parameters, determine whether the selectivity of the conversion of polycyclic aromatic hydrocarbons to n-alkanes meets expectations, and obtain the adjusted reaction pathway plan.

[0036] Reaction condition information is cleaned and normalized using data preprocessing techniques to generate a preliminary, organized efficiency dataset. Based on this efficiency dataset, a support vector machine algorithm is used to model and analyze the correlation between the degree of decycling and production efficiency, determining the trend of the degree of decycling. If the trend of the degree of decycling deviates from a preset threshold, the optimized pressure parameter configuration scheme is obtained by adjusting the simulated data input of the pressure parameters. The changes in the efficiency dataset are verified against this configuration scheme to determine the effectiveness of the decycling adjustment.

[0037] For example, when cleaning and normalizing reaction condition information, data preprocessing techniques can be used to remove outliers and missing values ​​to ensure data integrity and consistency. For example, in a hydrogenation reaction dataset, significant deviations were found in some temperature records. For example, the temperature value suddenly jumped to 200°C during a certain reaction period, while the normal range should be between 150-180°C. In this case, the abnormal data can be corrected through interpolation, and all data can be normalized and mapped to a range of 0 to 1 for subsequent analysis. This approach can effectively improve data availability and lay the foundation for subsequent modeling.

[0038] For example, using a support vector machine (SVM) algorithm to model and analyze the correlation between the degree of decycling and production efficiency in an efficiency dataset can be understood as using machine learning to identify the nonlinear relationship between the two. Suppose, in an experiment, the degree of decycling gradually increases from an initial 30% to 50%, while the production efficiency shows a trend of first increasing and then decreasing, reaching its peak at 40%. The SVM model can identify the key inflection points in this trend and provide a basis for subsequent parameter adjustments. This modeling approach helps to accurately grasp the relationship between key variables in the reaction process.

[0039] For example, to determine whether the trend in the degree of ring loss deviates from a preset threshold, a reasonable range can be set, such as 35%-45%. If the degree of ring loss reaches 48% during a given monitoring session, it indicates that the threshold has been exceeded and parameter adjustment is required. In this case, by simulating changes in pressure parameters, for example, gradually adjusting the pressure from 10 MPa to 12 MPa, the degree of ring loss can be observed to determine whether it falls back into the target range. This method allows for timely detection of deviations and the implementation of targeted measures.

[0040] For example, to optimize the pressure parameter configuration scheme, different pressure values ​​can be input through multiple simulations to obtain the corresponding response data.

[0041] For example, at a pressure of 11 MPa, the degree of ring removal dropped to 43%, and at 12 MPa, it further dropped to 41%. Combined with the changes in the efficiency data set, 11 MPa can be determined to be the optimal configuration. This verification method ensures the rationality of parameter adjustments and provides data support for process optimization.

[0042] For example, when validating changes in an efficiency dataset to determine the effectiveness of ring-debonding adjustments, you can compare the production efficiency data before and after the adjustments. For example, if the production efficiency was 75% before adjusting the pressure parameters and increased to 78% afterward, and the ring-debonding level remained stable within the target range, this indicates a significant improvement. This validation process helps confirm the actual impact of parameter optimization and accumulates experience for subsequent process improvements.

[0043] For example, from a broader perspective, the above approach can be extended to adjust other reaction conditions, such as flow rate or catalyst dosage, through similar data cleaning, modeling, and verification processes to comprehensively optimize the reaction system. This expansion approach can further improve the adaptability and stability of the process, providing more possibilities for industrial application.

[0044] S105. According to the adjusted reaction pathway plan, the hydrogenation process is run in the actual reactor, and the fluctuation range of the cetane number is extracted from the real-time monitoring data to determine whether the compatibility of the process conditions and the catalyst grading meets the target requirements.

[0045] A real-time monitoring system captures data on changes in cetane number during the hydrogenation process from within the reactor to generate an initial data set, and quantifies the distribution characteristics of the fluctuation range. If the distribution characteristics of the fluctuation range exceed a preset threshold, a data analysis module compares the process condition parameters to determine whether there are any abnormalities. If there are any abnormalities in the process condition parameters, a pre-established regression model is used to calculate the impact weight of the catalyst grading and determine the adjustment recommendation data. Based on the adjustment recommendation data, the degree of compatibility between the process conditions and the catalyst grading is analyzed.

[0046] For example, when obtaining data on changes in the cetane number through a real-time monitoring system during the hydrogenation process, attention can be paid to the fluctuations in the values ​​at different time periods within the reactor. Suppose that in a certain monitoring session, the initial data set shows that the fluctuation range of the cetane number within 24 hours is 10 to 15 units, while the preset threshold is 8 to 12 units, which is obviously beyond the expected range. In this case, it is necessary to further analyze the distribution characteristics of the fluctuation range, which may be caused by unstable reaction temperature or hydrogen flow. Through data analysis, it can be found that the fluctuations are mainly concentrated in the early stage of the reaction, which may be due to insufficient preheating of the raw materials or suboptimal catalyst activity.

[0047] For example, when comparing process parameters, the data analysis module can extract key indicators such as temperature, pressure, and hydrogen flow rate to determine if there are any anomalies. For example, if, during an experiment, the hydrogen flow rate suddenly drops to 70% of its normal value during a certain period, while all other parameters are within normal range, this abnormal flow rate can be preliminarily determined to be the primary cause of the excessive fluctuation. This analysis method helps quickly identify the root cause of the problem and provides a basis for subsequent adjustments.

[0048] For example, after identifying an outlier, using a pre-established regression model to calculate the impact weight of catalyst grading, one can assume that the model analysis results show that the proportion of a certain catalyst component has an impact weight of up to 0.6 on cetane number fluctuations, while the impact weight of other components is only around 0.2. This suggests that adjusting the proportion of this component may be the key to optimizing the process. Furthermore, through historical data comparison, it can be found that when the proportion of this component increases by 5%, the fluctuation range can be reduced to within the threshold, providing data support for adjustment recommendations.

[0049] For example, when analyzing the compatibility between process conditions and catalyst grading, an assessment can be conducted in conjunction with specific experimental scenarios. For example, if the current process conditions are high temperature and high pressure, while the catalyst grading is more suitable for medium temperature and medium pressure environments, the compatibility is poor, potentially leading to decreased reaction efficiency. By adjusting the recommended data, reducing the pressure by 10% and optimizing the ratio of active components in the catalyst, the cetane number fluctuation range can be observed to gradually stabilize. This matching analysis helps ensure the synergy between process conditions and catalyst performance, thereby improving overall reaction stability.

[0050] For example, from another perspective, the application of real-time monitoring systems can also be combined with historical data for trend prediction. For example, suppose an analysis of data from the past month reveals that the cetane number fluctuation range tends to increase at the beginning of each week. This could be due to equipment maintenance cycles or raw material batch variations. In this case, process parameters can be adjusted in advance or raw material batches can be changed to avoid excessive fluctuations. This predictive analysis can effectively reduce the probability of abnormalities.

[0051] For example, when implementing catalyst grading adjustment recommendations, the adjustment range can be further refined. Assuming the recommended data indicates that the proportion of a certain component needs to be increased by 3% to 5%, a small-scale test can be conducted initially with a 3% increase to observe whether the cetane number fluctuation improves. If the effect is not significant, the adjustment can be gradually increased to 5%. This step-by-step adjustment approach can reduce experimental risks while ensuring the accuracy of the adjustment direction. Through the above multi-angle analysis and examples, it can be seen that the complete process from data monitoring to parameter adjustment to matching degree assessment can effectively address potential problems in the hydrogenation process and provide a reliable basis for process optimization.

[0052] S106. If the fluctuation range of the cetane number exceeds a preset threshold, the process conditions are optimized using a gradient descent algorithm through regression analysis of the interactive effects of temperature control and pressure regulation to obtain a final process parameter combination.

[0053] By collecting cetane number data during the production process, the fluctuation range of this data is monitored in real time to determine whether it exceeds a preset threshold and obtain a preliminary fluctuation status assessment result. If the assessment result shows that the preset threshold is exceeded, the relevant records of temperature control and pressure adjustment are obtained from historical data, and regression analysis is used to explore the interaction and determine the main combination of factors affecting the fluctuation. Based on this combination of factors, a gradient descent algorithm is used to optimize the process conditions based on the interaction data of temperature control and pressure adjustment to obtain an adjusted set of process parameters.

[0054] For example, during the hydrogenation process, real-time cetane number data collected during production can effectively monitor whether its fluctuation range is within expectations. For example, during a production run, the system detected a short-term fluctuation of the cetane number from 42.5 to 48.5, exceeding the preset threshold range of 42.0 to 46.0. Initial assessment results indicate an abnormal fluctuation, necessitating further analysis. This monitoring approach can promptly identify potential issues and provide data support for subsequent adjustments.

[0055] For example, historical data can be used to extract relevant records of temperature control and pressure regulation when fluctuations exceed thresholds. Suppose historical records show that during an abnormal fluctuation, the reactor temperature suddenly increased from 320°C to 335°C, while the pressure dropped from 5.2 MPa to 4.8 MPa. By comparing multiple sets of data, it is found that the combination of increased temperature and decreased pressure may have caused the cetane number fluctuation to increase. This historical data mining helps quickly identify the influencing factors and provides a basis for process optimization.

[0056] For example, after determining that the interaction between temperature and pressure is the primary influencing factor, regression analysis can be used to further explore the specific weighting of their influence. Hypothetically, the analysis results indicate that temperature variations contribute 60% to cetane number fluctuations, while pressure variations contribute 40%. This analysis method clearly reveals the importance of each factor, helping process engineers focus on key parameters for adjustment, thereby improving control accuracy.

[0057] For example, when optimizing process conditions, a gradient descent algorithm can be used to find the optimal combination of temperature and pressure. Assuming an initial temperature of 330°C and a pressure of 5.0 MPa, multiple iterative adjustments ultimately resulted in a new optimized parameter set of 325°C and 5.3 MPa. During simulations, the cetane number fluctuation range narrowed to 43.0 to 45.5, meeting the preset threshold. This optimization method can effectively improve process stability and ensure controllability of the production process.

[0058] For example, if the interaction between temperature and pressure is complex, historical fluctuation data can be combined to analyze parameter trends across different production batches. For example, suppose one batch exhibits minimal fluctuation at 320°C and 5.1 MPa, while another batch exhibits good stability at 328°C and 5.4 MPa. By comparing these data, the rationale for the optimized parameters can be further verified. This multi-faceted verification approach helps enhance the reliability of the adjustment plan.

[0059] For example, for the optimized process parameter set described above, cetane number fluctuations can be continuously monitored during actual production to verify the effectiveness of the adjustments. If the fluctuations remain within the preset threshold over three consecutive days of operation, this indicates that the new parameter combination is well-suited to the catalyst grading. This continuous verification ensures long-term production and accumulates experience for subsequent process improvements.

[0060] S107. Run the full hydrogenation reaction using the final process parameter combination and catalyst grading scheme, extract the normal alkane content and cetane number from the product analysis data, and determine the precise conversion results of polycyclic aromatic hydrocarbons to high-cetane number alkanes.

[0061] Based on a pre-established database of process parameters, an optimized combination scheme for the hydrogenation reaction is obtained and, combined with the catalyst grading data, an initial operational configuration scheme is generated. Based on the monitoring results of the dynamic change trend of the reaction, a support vector machine model is used to extract key influencing factors and determine whether the matching degree between the process parameters and the catalyst grading meets a preset threshold. If the matching degree falls below the preset threshold, the parameter combination is adjusted and the operational configuration scheme is regenerated. The hydrogenation reaction is then executed using the adjusted scheme, and the chemical composition data of the final product is obtained to determine the conversion effect.

[0062] For example, during the process optimization of hydrogenation reactions, a pre-established process parameter database can be used to quickly identify parameter combinations that best suit the current feedstock characteristics. For example, if the database contains reaction temperature and pressure data for different feedstock viscosities and sulfur contents, for a batch of feedstock with a viscosity of 5.2 centipoise and a sulfur content of 0.8%, the historically optimal combination of 180°C temperature and 4.5 MPa pressure can be extracted as the initial solution. This process relies on extensive experimental records in the database to ensure the rationality of parameter selection.

[0063] Specifically, the application of catalyst grading data can further refine the initial operating configuration. Catalyst grading typically involves the proportions of different active components. For example, for high-sulfur feedstocks, a catalyst combination with a nickel-to-molybdenum ratio of 1:2 might be selected to improve desulfurization efficiency. By matching this grading data with process parameters, a configuration that synergizes temperature, pressure, and catalyst performance can be developed, providing a foundation for subsequent reactions.

[0064] For example, when monitoring reaction dynamics, support vector machine models can be used to identify key influencing factors. For example, if temperature fluctuations during a reaction cause a decrease in yield, the model, by analyzing historical data, may identify the interaction between temperature fluctuations and hydrogen flow rate as the primary influencing factor. This approach can extract core variables from multi-dimensional data, helping to pinpoint the root cause of the problem.

[0065] In one embodiment, when determining whether the matching degree between process parameters and catalyst grading reaches a preset threshold, the matching degree threshold can be set to 85%. If data comparison reveals that the matching degree of the current combination is only 78%, the parameter combination needs to be adjusted, for example, by slightly adjusting the pressure from 4.5 MPa to 4.7 MPa to improve the reaction activity and the synergistic effect of the catalyst. After the adjustment, the operation configuration plan is regenerated to ensure that the reaction conditions are more in line with actual needs.

[0066] Specifically, after executing the adjusted scheme for hydrogenation reaction, it is particularly important to obtain the chemical composition data of the final product.

[0067] For example, gas chromatography analysis may reveal that the sulfur content in the product has dropped from 0.8% to 0.05%, indicating a significant conversion effect. This data analysis provides a direct basis for subsequent process improvements and also verifies the effectiveness of parameter adjustments.

[0068] For example, from another perspective, parameter adjustments and catalyst grading optimization can also be supplemented from the perspective of raw material adaptability. To account for minor differences between raw material batches, the database can store multiple backup plans. For example, for raw materials with slightly higher sulfur content, the molybdenum content in the catalyst could be increased to 60%, and the reaction temperature could be appropriately lowered to 175°C to avoid side reactions. This flexible adjustment approach can enhance process adaptability.

[0069] In one embodiment, the coordinated efforts of these various steps not only ensure the stability of the hydrogenation reaction but also effectively improve the quality consistency of the products. Through dynamic monitoring and parameter optimization, the process can rapidly respond to changes in feedstock or the environment, ensuring production efficiency while reducing the proportion of substandard products. This comprehensive optimization approach provides reliable technical support for industrial production.

[0070] S108. Obtain long-term operating data for accurate conversion results, analyze the stability of catalyst grading and process conditions through machine learning algorithms, determine the dynamic adjustment requirements of the reaction path, and obtain an adaptive hydrogenation optimization model.

[0071] Based on the fluctuation data of the catalyst gradation and process conditions, statistical tools are used to extract features and determine the distribution characteristics of key influencing factors. Based on these distribution characteristics, the correlation between the catalyst gradation and the process conditions is modeled using a random forest algorithm to determine the potential change trend of the reaction pathway. If the change trend of the reaction pathway exceeds a preset threshold, parameter screening is performed on the process conditions to obtain a set of candidate solutions for dynamic adjustment. Based on this set of candidate solutions and combined with historical data evaluation, the optimal adjustment strategy is determined.

[0072] For example, when analyzing fluctuating data on catalyst grading and process conditions, statistical tools can be used to extract features from the data, focusing on key variables such as reaction temperature, pressure, and the proportion of active catalyst components. For example, in a hydrogenation reaction, the temperature fluctuates between 350 and 380 degrees Celsius and the pressure fluctuates between 8 and 10 MPa. Statistical tools can help extract the distribution characteristics of these variables, such as a mean temperature of approximately 365 degrees Celsius and a standard deviation of 5 degrees Celsius, to determine whether these fluctuations have a significant impact on the reaction path. This analysis of distribution characteristics provides the data foundation for subsequent modeling.

[0073] For example, when using the random forest algorithm to model the correlation between catalyst grading and process conditions, the content ratio of active metals in the catalyst can be used as an input variable, while process conditions such as reaction time and hydrogen flow rate can be used as auxiliary variables.

[0074] In one possible implementation, assuming the nickel content in the catalyst is adjusted from 5% to 7% and the hydrogen flow rate is increased from 200 liters / hour to 250 liters / hour, the model can predict the impact of this combination on the PAH conversion rate. By analyzing the model output, it can be determined whether the reaction path deviates from the expected path, such as whether more byproducts will be generated. This modeling approach helps to reveal the complex relationships between variables and provides theoretical support for process optimization.

[0075] For example, when determining potential changes in a reaction pathway, if the model's predictions indicate a deviation from a preset threshold, such as a 3 percentage point drop in n-alkanes yield, historical data comparison can confirm that this is due to an increase in side reactions caused by excessively high temperatures. At this point, parameter screening is performed on the process conditions to generate a set of candidate solutions, such as lowering the temperature to 360 degrees Celsius or fine-tuning the pressure to 9.5 MPa. These candidate solutions provide multiple possibilities for subsequent adjustments, ensuring that the reaction pathway returns to its intended path.

[0076] For example, when evaluating and determining the optimal adjustment strategy based on historical data, a combination of a temperature of 360 degrees Celsius and a pressure of 9.5 MPa can be selected from the candidate set. By comparing historical operating data, it was found that this combination, under similar fluctuating conditions in the past, was able to increase the cetane number by approximately 2 units while maintaining product stability. This evaluation method not only improves the reliability of the adjustment strategy but also effectively avoids the risks associated with improper parameter adjustments. Ultimately, this data-driven dynamic adjustment method can better adapt to the uncertainties in the reaction process and improve the adaptability and stability of the overall process.

[0077] The description of the above embodiments is only used to help understand the technical solutions and core ideas of this application; ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for accurately converting coal tar into high cetane number alkanes by hydrogenation, characterized in that: The method comprises: Obtain chemical composition data of coal tar samples, analyze the distribution ratios of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons, and alkanes, determine the characteristic parameters of the initial reactants, and obtain benchmark values ​​for the proportion of polycyclic aromatic hydrocarbons and the target conversion pathway; By optimizing the catalyst grading scheme, combined with temperature control and pressure regulation parameters, a process condition simulation system was constructed to calculate the dynamic changes of hydrogenation depth and ring-opening equilibrium in the virtual reactor and determine the preliminary process condition range; Acquire reaction data within the initial process conditions. If the hydrogenation depth exceeds a preset threshold, adjust the temperature control parameters to reduce the degree of decyclization. A stable n-alkane production efficiency is obtained through iterative calculation. Focusing on the stable n-alkanes production efficiency, the changing trend of the decyclization degree in the reaction pathway is analyzed. By optimizing the pressure adjustment parameters, it is determined whether the selectivity of the conversion of polycyclic aromatic hydrocarbons to n-alkanes meets the expectations, and the adjusted reaction pathway scheme is obtained. Based on the adjusted reaction pathway, the hydrogenation process was run in an actual reactor. The cetane number fluctuation range was extracted from the real-time monitoring data to determine whether the process conditions and catalyst grading were compatible with the target requirements. If the cetane number fluctuation range exceeds the preset threshold, the interactive effects of temperature control and pressure regulation are analyzed through regression, and the gradient descent algorithm is used to optimize the process conditions to obtain the final process parameter combination; Using the final process parameter combination and catalyst grading scheme, the full hydrogenation process was run, and the n-alkane content and cetane number were extracted from the product analysis data to determine the precise conversion of polycyclic aromatic hydrocarbons to high-cetane number alkanes. Obtain long-term operating data for accurate conversion results, analyze the stability of catalyst grading and process conditions through machine learning algorithms, determine the dynamic adjustment needs of the reaction path, and obtain an adaptive hydrogenation optimization model.

2. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The chemical composition data of the coal tar sample is obtained, and the distribution ratio of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons and alkanes is analyzed to determine the characteristic parameters of the initial reactants, thereby obtaining the reference values ​​of the polycyclic aromatic hydrocarbon content ratio and the target conversion path, including: Detecting the coal tar sample using spectral analysis technology to obtain original distribution data of polycyclic aromatic hydrocarbons, monocyclic aromatic hydrocarbons and alkanes in the sample; Calculating the distribution ratio of each component based on the original distribution data, and determining the relative content ratio of each component using a preset classification model; If the content ratio of the polycyclic aromatic hydrocarbons exceeds a preset threshold, the conversion paths are prioritized through the information processing step to obtain a preliminary solution; Based on the preliminary plan and in combination with the alkane distribution data, the direction of the pathway adjustment is determined to determine the final target conversion pathway benchmark value.

3. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The optimized catalyst grading scheme is combined with temperature control and pressure adjustment parameters to construct a process condition simulation system, calculate the dynamic changes of hydrogenation depth and ring-opening equilibrium in the virtual reactor, and determine the preliminary process condition range, including: By building a simulation system, we can obtain the reaction data of the catalyst grading scheme under different temperature and pressure parameters and determine the preliminary dynamic change trend; According to the dynamic change trend, a virtual reactor is used to simulate the hydrogenation depth and the ring-opening equilibrium in real time to obtain the key fluctuation range in the reaction process; For the key fluctuation range, a control combination of the temperature parameter and the pressure parameter is obtained, and a pre-established regression model is used to analyze the impact of the control combination on the reaction simulation, and the optimal control range is determined to optimize the reaction process.

4. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The reaction data within the preliminary process conditions are obtained, and if the hydrogenation depth exceeds a preset threshold, the temperature control parameters are adjusted to reduce the degree of decyclization, and a stable normal alkane production efficiency is obtained through iterative calculation, including: Obtaining initial reaction data from process conditions, and standardizing the reaction data using a data processing tool to obtain a preliminary organized reaction data set; Based on the reaction data set, real-time monitoring is performed on the hydrogenation depth. If the hydrogenation depth exceeds a preset threshold, a condition adjustment mechanism is triggered to determine the adjustment range of the temperature parameter; The real-time change data of the degree of ring-off are obtained through the adjusted temperature parameter range, and the change data are fitted using a regression analysis model to determine whether the degree of ring-off tends to be stable.

5. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The method aims to stabilize the efficiency of normal alkane production, analyze the changing trend of the degree of decyclization in the reaction pathway, and determine whether the selectivity of the conversion of polycyclic aromatic hydrocarbons to normal alkanes meets expectations by optimizing the pressure adjustment parameters, thereby obtaining an adjusted reaction pathway scheme, including: The reaction condition information is cleaned and normalized through data preprocessing technology to obtain a preliminary organized efficiency data set; Based on the efficiency data set, a support vector machine algorithm is used to model and analyze the correlation between the degree of de-looping and the generation efficiency to determine the changing trend of the degree of de-looping; If the change trend of the ring-out degree deviates from the preset threshold range, the optimized pressure parameter configuration scheme is obtained by adjusting the simulated data input of the pressure parameter; With respect to the configuration scheme, the change of the efficiency data set is verified to determine the adjustment effect of the de-loop degree.

6. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The method includes running a hydrogenation process in an actual reactor according to the adjusted reaction pathway scheme, extracting the cetane number fluctuation range from real-time monitoring data, and determining whether the compatibility of process conditions and catalyst grading meets target requirements, including: Acquiring cetane number change data during hydrogenation from the reactor through a real-time monitoring system to obtain an initial data set and quantify the distribution characteristics of the fluctuation range; If the distribution characteristics of the fluctuation range exceed the preset threshold, the process condition parameters are compared through the data analysis module to determine whether there are abnormal points; If there are abnormal points in the process condition parameters, a pre-established regression model is used to calculate the influence weight of the catalyst grading to determine the adjustment suggestion data; Based on the adjustment suggestion data, the matching degree between the process conditions and the catalyst grading is analyzed.

7. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: If the cetane number fluctuation range exceeds the preset threshold, the interactive effects of temperature control and pressure regulation are analyzed through regression analysis, and the gradient descent algorithm is used to optimize the process conditions to obtain the final process parameter combination, including: By collecting cetane number data during the production process, real-time monitoring of the fluctuation range of the data is performed to determine whether it exceeds a preset threshold and obtain preliminary fluctuation status assessment results; If the evaluation result shows that the preset threshold is exceeded, relevant records of the temperature control and pressure regulation are obtained from historical data, and regression analysis is used to explore the interaction and determine the main factor combination affecting the fluctuation; According to the combination of factors, a gradient descent algorithm is used to optimize process conditions for the interaction data of the temperature control and pressure regulation to obtain an adjusted process parameter set.

8. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The final process parameter combination and catalyst grading scheme are used to run the full hydrogenation reaction process, extract the normal alkane content and cetane number from the product analysis data, and determine the precise conversion results of polycyclic aromatic hydrocarbons to high cetane number alkanes, including: Obtaining an optimized combination plan for the hydrogenation reaction based on a pre-established process parameter database, and generating an initial operation configuration plan in combination with the catalyst grading data; Based on the monitoring results of the dynamic change trend of the reaction, a support vector machine model is used to extract key influencing factors to determine whether the matching degree between the process parameters and the catalyst grading reaches a preset threshold; If the matching degree is lower than a preset threshold, adjusting the parameter combination and regenerating the operation configuration scheme; The hydrogenation reaction is performed using the adjusted scheme to obtain chemical composition data of the final product and determine the conversion effect.

9. The method for accurately converting coal tar into high cetane number alkanes by hydrogenation according to claim 1, characterized in that: The method obtains long-term operating data of accurate conversion results, analyzes the stability of catalyst grading and process conditions through machine learning algorithms, determines the dynamic adjustment requirements of the reaction path, and obtains an adaptive hydrogenation optimization model, including: Based on the fluctuation data of the catalyst gradation and process conditions, statistical tools are used to extract features and determine the distribution characteristics of key influencing factors; Based on the distribution characteristics, the correlation between the catalyst gradation and the process conditions is modeled using a random forest algorithm to determine the potential change trend of the reaction pathway; If the change trend of the reaction path exceeds a preset threshold, parameter screening is performed on the process conditions to obtain a set of dynamically adjusted candidate solutions; The optimal adjustment strategy is determined by combining the candidate solution set with historical data evaluation.

Images