Document Type : Original Article

Highlights

Fuel economy is one of the most important challenges for the petrochemical industry sector as it faces a shortage of fossil fuels and rising prices. A significant portion of the country's energy consumption in process units such as refineries, power plants, and petrochemical complexes is used as heat in furnaces and burners, which usually have a thermal efficiency of less than 60%. Therefore, fuel economy and optimal use of energy in these sectors can play a decisive role in reducing operating costs and controlling environmental pollutants. Most studies conducted on reformer furnaces have focused on optimizing fuel consumption, reducing pollutants, and increasing thermal performance [1-12].

In recent years, several studies have been conducted in the field of optimizing fuel consumption and improving thermal efficiency in industrial furnaces. Changing the fuel composition and examining its effect on combustion characteristics and pollutants has been one of the main axes of Samai et al. [13]. In this study, the performance of a high-speed burner connected to the furnace was numerically investigated using ANSYS software and the effect of changing the fuel from methane to propane on combustion parameters was analyzed. The results showed that despite the reduction in the volumetric flow rate of the fuel, the flame structure and combustion characteristics change in such a way that the burner performance remains within the range of high-speed burners and at the same time, the production rate of nitrogen oxide (NOx) with propane fuel increases but remains in the low-NOx range. These results indicate the importance of the type of fuel and its composition in controlling the combustion quality and thermal efficiency of furnaces.

So far, the impact of fuel changing in furnaces using surplus unit fuels has rarely been studied. Mirvakili et al. [20] previously investigated the effect of fuel changing in the reformer furnace on the thermal performance of the furnace, without considering the tubes and their impact on the syngas reaction. The best thermal performance of the furnace occurs when the amount of hydrogen in the replacement fuel is lower. They found that a 64% reduction in hydrogen content in the furnace fuel led to a 10% increase in flame length and a 14% increase in the share of radiative heat transfer. However, the effect of fuel composition changes on the syngas reaction inside the tubes has not yet been studied. In this study, the main objective is to investigate the impact of fuel switching on the chemical reaction performance of methane reforming.

Geometry and boundary conditions

This furnace is of the high-fire type and its dimensions are approximately 5 m wide, 15 m long and 12 m high. In the reaction section, 184 vertical tubes with a height of about 12 m and an internal diameter of 12 cm are installed, through which the input feed, consisting of a mixture of natural gas and water vapor, passes. Due to its highly endothermic nature, the steam reforming reaction requires a significant supply of heat, which is provided by the combustion of fuel in 70 burners embedded in the walls. The geometry of the reformer is shown in Table 1.

The furnace consists of a total of 5 tunnels, in which about 750 outlet openings are installed to discharge hot combustion gases. Such a design allows for better control of the flow of hot gases, reducing pressure drop and increasing the thermal efficiency of the furnace. Due to the geometric symmetry of the furnace and in order to reduce the volume of calculations, only half of the furnace, consisting of 92 tubes and half of the burners, has been simulated in this study.

In this study, six different scenarios for the fuel consumed by the reformer furnace have been investigated. The first scenario represents the actual operating conditions that, after about ten years of operation, have gradually deviated from the initial design conditions (scenario six). The other five scenarios have been introduced as proposed cases that are compared with the operating and design conditions. These scenarios have been defined based on the use of side streams and waste gases available in the petrochemical complex.

The gases used in the proposed cases include:

Purge gas: Part of the unreacted gas flow in the methanol synthesis reactor that is removed from the process cycle.

Expansion gas: The flow that is released after generating electrical power in the expansion turbine.

OFF gas: The outlet flow of the hydrogen separation unit from unreacted gases.

All six cases used in this study are presented in Table 2. Compared to the design case (sixth), in the operating case (first), the total fuel consumption including natural gas and purge gas has increased. In the second scenario, compared to the operating case, natural gas consumption is reduced and part of the energy required is provided by expansion gas. In the third scenario, natural gas consumption is reduced to the design level and the share of expansion gas in energy supply has increased. In the fourth scenario, the natural gas consumption is similar to the second case, with the difference that part of the purge gas is reduced and expansion gas is replaced by it. In the fifth scenario, the entire purge gas flow is eliminated and a combination of expansion gas and excess gas is used as an alternative fuel. It should be noted that the minimum calorific value of fuels in all scenarios is approximately within the same range, and the air required for combustion in each case was calculated with an additional 20%, and the results are presented in Table 2.

Results and Discussion

To validate the numerical model, simulation results were compared with design and operational data. As shown in Table 5, the temperature difference between the furnace and the tubes in the simulation is very small compared to the experimental values, with an absolute error of less than 1%. Additionally, the comparison of the outlet gas composition shows that the molar fractions of hydrogen, methane, carbon dioxide, and carbon monoxide are in good agreement with the reference data, with the maximum error limited to a few percent. These results indicate a high accuracy of the numerical model in reproducing real process conditions and its capability to analyze the behavior of the reformer furnace under various operational conditions.

Table 6 shows the comparison of simulation outputs 1 to 6. As can be observed, the outputs of the different cases do not show significant differences; this is because the selected fuels have almost the same lower heating value. The temperature difference in the furnace exhaust gas is about ±5°C, and in the best case, hydrogen production has increased by up to 5%. The comparison of the fuel flow rate entering the burners (Table 6) indicates that scenarios 6, 3, 4, and 5 consume less natural gas compared to the operational case (Scenario 1). The highest reduction in gas consumption is observed in scenarios 6 and 5; such that in scenario 5, natural gas consumption is 16% less than the operational case, while hydrogen production has increased by about 2%. Although the exhaust gas temperature in scenario 5 is slightly lower than in case 1, an increase in hydrogen production and a decrease in residual methane are simultaneously observed. The reason for this behavior can be seen in the analysis of the profiles related to hydrogen production, the gas temperature inside the tubes, and the tube surface temperature, which are presented in the following.

Figure 6a shows the methane profile inside the reformer tube. In scenario 5, methane is consumed more rapidly in the first four meters due to the increased share of radiative heat transfer in this region; this behavior is also observed in the temperature profile. Overall, only scenarios 4 and 5 show better performance than the operational case, with higher methane consumption. In scenario 6 (design condition), due to higher emissivity coefficients, the highest methane consumption is observed. Figure 6b shows the molar fraction profile of hydrogen in different cases. In scenario 5, hydrogen production increases significantly in the first four meters and then remains almost constant along the rest of the tube. The reason is the supply of energy required for the endothermic steam methane reforming reaction at the beginning of the tube and the system reaching equilibrium conditions further along the path. In other scenarios, the hydrogen production slope in the first four meters is lower and the hydrogen concentration increases gradually until the end of the tube. Figures 6c and 6d show the molar fraction profiles of carbon monoxide and carbon dioxide, respectively. The variations of these two components in different scenarios remain within about 0.01 molar fraction, and no significant differences are observed between them.

Conclusions

In this study, the effect of changing the fuel composition of the methanol unit reformer furnace on the thermal performance of the furnace and the behavior of the reforming reactions inside the tubes was investigated. Six different scenarios, including the operational condition, the design condition, and four proposed cases based on the use of by-product gases of the unit (blowdown gas, expansion gas, and surplus gas), were simulated and compared.

The results showed that due to the similarity in the lower heating value of the fuels, the temperature of the furnace exhaust gas remains within a narrow range (about ±5 K) in all scenarios. However, the fuel composition has a significant effect on the flame length, the heat distribution between radiative and convective mechanisms, and ultimately on the temperature and gas composition inside the tubes. Scenarios with higher hydrogen content (such as 2 and 3) caused an increase in the tube surface temperature in the lower region and a shortening of the flame length. This phenomenon occurs due to the high reactivity of hydrogen and the dominance of convective heat transfer, which has an adverse effect on the distribution of the reforming reaction along the tube. In contrast, the increase in methane share (Scenario 5) led to an increase in flame length and the expansion of radiative heat transfer, which increased methane consumption and hydrogen production in the initial section of the tubes.

In terms of energy efficiency, the results showed that natural gas consumption in scenarios 4, 5, and 6 decreased significantly. Particularly in scenario 5, a 16% reduction in natural gas consumption was recorded along with a 2% increase in hydrogen production. This indicates that the substitution of methanol unit by-product gases can not only reduce the consumption of the main feedstock but also improve the performance of the reforming process. Finally, the comparison of the design condition (Scenario 6) with other cases showed that the role of the furnace emissivity coefficients in determining thermal conditions is much more significant than the fuel composition change; such that even with similar compositions, an increase in emissivity coefficients can lead to higher temperature and methane consumption.

In general, it can be concluded that the use of by-product gases in the fuel composition of the reformer furnace not only provides the possibility of reducing natural gas consumption but can also lead to an improvement in hydrogen production efficiency. However, the selection of the optimal composition should be made by considering the balance between flame length, heat transfer mechanisms, and reforming reaction conditions to ensure both cost reduction and process stability and efficiency.

Subjects

• طراحی و شبیه سازی کوره ها