Method for improving temperature homogenization in a steam methane reformer by adjusting the power distribution

Abstract

A method for improving an endothermic process in a furnace comprising reforming tubes positioned vertically in rows within the furnace, wherein the inner row of burners is positioned between two rows of tubes and the outer row of burners is positioned between a row of tubes and a side wall parallel to the row of tubes, wherein the method improves temperature homogenization of the tubes by throttling a portion of the burners, the method comprising the steps of a) optionally calibrating a simplified physical model of step c3) by measuring one or more actual tube temperatures of at least one tube affected by burner throttling in standard and throttled states, b) obtaining information about the actual tube temperatures of the tubes present in the furnace, all burners present in the furnace being under standard non-throttling conditions, c) obtaining a map of the burners to be throttled, comprising c1) selecting at least one parameter representative of the performance of the furnace and a target for improvement, c2) selecting at least one or more power ratios for burner throttling, c3) utilizing the information of step b) and a simplified physical model of the effect of throttling the burners on tube temperature, c4) obtaining a map of the burners to be throttled, and d) throttling the burners.

Description

[0001] The invention relates to a method for improving an endothermic process taking place in a furnace comprising tubes filled with a catalyst for the chemical conversion of a gaseous feed and positioned vertically in rows within the furnace, the burners being mounted in rows, wherein the inner burners of a row are positioned between two rows of tubes and the outer burners of a row are positioned between a row of tubes and a side wall parallel to the row of tubes, wherein the heat transfer from the burners to the tubes is correlated with the flow rate, wherein the method improves the temperature homogenization of the tubes by throttling a portion of the burners.

[0002] The method allows for improved reformer operation in terms of safety and debottlenecking and is particularly applicable to top-fired or bottom-fired furnaces used for steam methane reforming (SMR) and other endothermic reactions such as cracking of hydrocarbon feedstocks in externally fired reactors.

[0003] The SMR process is primarily based on the reforming reaction of light hydrocarbons (such as methane) in the presence of steam to produce a mixture of hydrogen (H2) and carbon monoxide (CO). This reaction is endothermic and slow, and requires additional heat input and a catalyst to occur. Generally, the performance of an SMR reactor is limited by heat transfer rather than by the kinetics of the reaction.

[0004] In industrial practice, an SMR reactor generally comprises a tubular reformer (also called tube) placed in a furnace, the tube being filled with a catalyst (usually in pellet form) and fed with a process gas mixture of methane and steam.

[0005] Several well-proven configurations are available for furnace design: top-fired (also known as underfired), bottom-fired (also known as overfired), side-fired, and stepped wall.

[0006] Top-fire technology was one of the most mentioned designs and was proposed by several technology providers.

[0007] For simplicity, most of the following description is based on top-fired furnaces, however most of the figures and description are also applicable to bottom-fired furnaces. The description is based on steam methane reforming processes, but is also applicable to other endothermic reactions such as cracking of hydrocarbon feeds in externally fired reformers.

[0008] Top-fired furnaces typically consist of a refractory-lined furnace containing several rows of catalyst-containing tubes. For each row, syngas is collected from each tube into a common duct outside the furnace, known as the syngas collector. The syngas collected from all rows is fed into a crossover header and then passed to a process gas cooler.

[0009] The heat required for the endothermic reaction to occur is provided by rows of top burners placed between the tubes, as well as by additional rows of top burners at the sides of the furnace, along the sidewalls parallel to the tube rows. The combustion products from the burners are typically blown vertically downward, so that the upper portion of the tube rows face the flames. A flue gas collector is typically located at the level of the furnace floor. The outer row of burners, i.e., along the sidewalls, heats only one row of tubes on one side and the refractory sidewalls on the other. Each row of burners in the inner row of burners, located in the center of the furnace, heats two rows of tubes on either side of the burner row. Therefore, the outer burners need to provide less power than the inner burners.

[0010] Bottom-fired technology is less common in modern equipment. In this technology, the burners are also arranged in rows, but they are mounted on the floor of the combustion area between the tube banks and fire vertically upwards.

[0011] The primary goal of furnace (also known as hearth) design is to maximize the heat transferred from the burner to the tube (from the burner flame as well as from the walls and hot flue gases) while complying with the tube maximum operating temperature limit. The tube maximum operating temperature or MOT (also known as the maximum operating limit) varies with several factors, particularly the tube mechanical load (primarily the feed gas pressure), the mechanical properties of the alloy used for the tube, and the expected life of the tube in terms of exposure to creep and thermal aging.

[0012] Any increase in heat transfer to the tubes has a direct positive impact by increasing productivity or by improving the compactness of the furnace (which has value in terms of capital expenditure). However, increased heat transfer generally means higher tube surface temperature (TST) levels, which can reduce tube life or require more expensive resistant alloys.

[0013] Lack of uniformity in the heat load distribution in the furnace will cause some tubes to run hotter than others, which is why the tube temperature profile is a key factor in furnace design and operation. When looking for a good compromise between performance and durability (a good compromise is indeed essential), the tube surface temperature TST curve (also known as tube wall temperature TWT or simply tube temperature) provides decisive information.

[0014] Therefore, during operation, the performance of the furnace is limited by the temperature of the hottest tube (i.e., the maximum tube temperature or MTT), which should not be hotter than the MTT. At the same time, the performance of the process (productivity or conversion efficiency, or a combination of both) depends on the average tube heat flux and temperature. Therefore, the smaller the difference between the hottest and coldest tube temperatures, the better the furnace performance.

[0015] In a common design, the lower ends of the tubes in a row are connected to syngas collectors. For each syngas collector, an associated temperature sensor measures the temperature of the collected syngas. This temperature (called the "syngas collector temperature") represents the average temperature of the connected tubes. The difference between the temperature of the hottest syngas collector and the temperature of the coldest syngas collector (called the "syngas collector temperature gap") is then used to provide an additional estimate of performance.

[0016] In steam methane reforming, the tubes contain a catalyst medium to enable the reforming reaction. In a top-fired furnace, the feed is supplied to the tubes at their top ends, and the resulting syngas (containing hydrogen and carbon monoxide as primary components, along with residues) is exhausted at the bottom. The furnace's combustion chamber is typically rectangular, with burners arranged in rows along the length of the furnace. Inner burners are arranged between the tube rows, and outer burners are arranged between the outer tube rows and the side walls to provide the heat required for the process gas conversion. Flue gases are extracted through an exhaust duct.

[0017] An inherent consequence of furnace design limitations is uneven transfer of available heat between the reformer tubes. Listed below are some of the causes of this uneven transfer:

[0018] - the way in which the tubes and burners are arranged in rows within the furnace;

[0019] - Uneven distribution of air and fuel flows from different manifolds into the burner nozzles;

[0020] - The jet momentum flux is different between the outer row burners and the inner row burners;

[0021] - Unbalanced feed gas distribution;

[0022] - offset of the passage opening for the flue gas outlet;

[0023] -Geometric effect of convergence of flue gas ducts.

[0024] Further differences between tube temperatures may occur in the field:

[0025] - During furnace construction, such as uneven catalyst filling or incorrect burner installation;

[0026] - In addition, during equipment development, such as burner failure issues, operating conditions that produce coke or catalyst aging on the burner gas tip or facilities.

[0027] All of these imperfections that are inherent in the furnace design, or that may arise during furnace construction or develop while the unit is in operation, result in a difference between the hottest and coldest tube temperatures, known as the "tube temperature gap."

[0028] The tube temperature gap must be reduced to ensure an optimal balance between efficiency and tube life in an SMR furnace.

[0029] Unanticipated discrepancies must also be corrected and / or compensated for. Therefore, finding mitigation strategies is of great importance for improving the operation of SMR plants.

[0030] When differences in tube behavior may not be prevented or foreseen, one solution is to apply a field-applicable remedial method that mitigates temperature differences between reformer tubes.

[0031] One solution to reduce the temperature differences is to adjust the burner power individually to make the heat transfer to the reformer uniform; this can be achieved by throttling the individual burners.

[0032] The resulting reduction in temperature differences will improve equipment performance; another advantage is that the solution can be applied during normal operation without having to wait for maintenance periods.

[0033] The burner is supplied with a primary and secondary fuel stream and an oxidant stream. Variation of the burner power can be obtained by acting on one or more of the streams:

[0034] - A secondary fuel stream (usually tail gas), which usually contributes most to the burner power;

[0035] - the main fuel stream (usually natural gas), which has less flexibility in terms of flow rate variations, since a minimum amount is required to ensure good burner operation in case of fluctuations in the composition of the main contributor;

[0036] - Oxidant flow; this is also theoretically possible, but this flow has the highest velocity, so it is necessary to pay attention to the interaction between burners with different momentum flux ratios.

[0037] Reduction in the flow rate of at least one of the fuel stream (primary or secondary) or the oxidant stream may be achieved by throttling its associated valve or by installing a flow restriction orifice.

[0038] Burners that reduce power by reducing the flow rate of one or several streams are called "throttling burners".

[0039] Alleviating the tube temperature gap, and therefore mitigation strategies, are of great significance for improving the operation of SMR equipment.

[0040] In SMR plants, it is now standard practice to equip the burners with valves so that throttling of the burners does not require shutting down the plant.

[0041] Reducing the difference in tube temperature by throttling the burners is known per se, however, the method can be greatly improved by judicious selection of a suitable burner group.

[0042] It is known to select a group of burners to be throttled by empirical trial and error. This method is practiced by those skilled in the art for existing equipment, however it is time consuming and the results may not be satisfactory.

[0043] Another solution is proposed in EP 2325562, which discloses a method for operating a furnace in which tube temperatures need to meet selected target temperature criteria. The method provides a systematic and quantitative approach to determining how to adjust the burner flow rate to produce the desired tube surface temperature, for example, to minimize the temperature deviation between tube wall temperatures at predetermined heights within the furnace. The burner flow rate is adjusted based on calculated target flow rates, which are calculated using an estimate of a mathematical function and temperature information. This method requires acquiring information about the temperature of the process tubes by capturing images comprising pixel data and processing the pixel data to obtain temperature information. The method also includes providing an estimate of a mathematical function that characterizes the relationship between changes in the burner flow rate and changes in the temperature of the process tubes. A drawback of this method is its complexity and the time required to estimate the parameters of the mathematical function, which are required from iterative TST measurements associated with acquiring and processing images comprising temperature information. Consequently, this method requires on-site materials, manpower, and time to operate an IR camera to evaluate the relationship between burner throttling and TST.

[0044] As mentioned above, finding the optimal burner for throttling can be tricky and time-consuming in practice, especially for large furnaces with a large number of burners and tubes, which is a real weakness of the known solutions.

[0045] Therefore, there is still a need for an improved throttling method that allows identifying a set of burners suitable for throttling and throttling them to mitigate the tube temperature differences. More precisely, there is a need for a method that allows obtaining a set of burners suitable for throttling, due to an efficient, fast and easy-to-implement method, thus offering definite advantages over known solutions.

[0046] The present invention aims to propose a method for improving and controlling the surface temperature of tubes present in a furnace during operation, by efficiently and quickly determining a suitable set of burners (also known as a burner map) whose power should be advantageously reduced in order to reduce temperature differences in the tubes of the furnace. Thanks to the solution of the present invention, a suitable burner map to be throttled in order to achieve a desired target for at least one selected parameter can be achieved more quickly than with known solutions. The purpose of selecting the parameter is to improve the performance of the process. Examples of selected parameters and associated targets are listed below:

[0047] - The parameter is the maximum TST and the goal is to reduce the maximum TST;

[0048] -The parameter is TST gap, and the goal is to reduce TST gap;

[0049] -The parameter is the syngas collector temperature gap, and the goal is also to reduce the SCT gap.

[0050] This method thus limits on-site labor requirements and the risks associated with them. Thanks to the present invention, on-site personnel will only need to perform one set of TST measurements and, if this task cannot be performed remotely from the control room, ultimately manually throttle the valve. For the first implementation of this method in a plant, a calibration step is required, which involves obtaining information about the actual tube temperature of at least one tube, preferably the tube closest to the burner, under both unthrottled and throttled conditions. This information can be obtained using existing on-site tube temperature measurement devices, typically using thermocouples and / or pyrometers and / or infrared cameras, but can also be obtained by any other suitable means.

[0051] Using the prior art solution to obtain the above-mentioned burner map to be throttled, the map takes too much time to obtain, and therefore it is difficult to use this solution frequently in the field for operators to make decisions in their daily work.

[0052] By applying the solution of the present invention, this disadvantage is avoided. Thanks to the invention, it is possible to determine in a few minutes the burner diagram to be throttled at one or more selected power ratios τ. The power ratio τ of a given burner represents the power delivered by this burner.

[0053] In the context of the present invention, the power ratio τ is:

[0054]

[0055] Among them, P 非节流 is the power of the burner in normal operating mode (i.e. nominal mode), and ΔP is the power change caused by throttling:

[0056] Or ΔP<0: burner power is reduced due to burner throttling;

[0057] Or ΔP>0: The situation where the burner power of the non-throttling burner increases due to the power redistribution among the non-throttling burners in the furnace.

[0058] The power ratio τ for a given burner can also be expressed as a percentage, in this case:

[0059] τ% = 100% means that the burner and all burners are in the standard (ie nominal) state and ΔP = 0

[0060] τ%=0% means the burner is closed; ΔP=P 非节流

[0061] τ% > 100% means the burner is in standard condition, but with power redistribution - this takes into account the throttling of other burners in the furnace. When at least one burner is throttled, the excess power is redistributed among the non-throttling burners to keep the total power of all burners constant.

[0062] τ% < 100% means the burner has been throttled to reduce the power delivered by this burner

[0063] The solution of the present invention relies on:

[0064] 1) The method first implements the required calibration step in the device. This step means obtaining information about the actual tube temperature of the individual tubes closest to the burner before and after throttling.

[0065] 2) obtaining initial data by acquiring information about the actual tube temperatures of the tubes present in the furnace under unthrottled conditions,

[0066] 3) Automatically obtain the burner map to be throttled by applying the following method:

[0067] i) selecting parameters and targets representing device performance - typically maximum TST and reduction, and / or TST gap and reduction, and / or SCT gap and reduction, and / or other suitable parameters related to the associated targets,

[0068] ii) selecting one or more power ratios to be applied to a plurality of burners in the furnace,

[0069] iii) then automatically calculating the changes in TST in response to the throttling of the different burners in the furnace to automatically obtain the best burner map to be throttled,

[0070] 4) Finally, the burners are individually throttled according to the burner map to be throttled.

[0071] The inventors have found that by applying some rules that simplify the calculation of the impact of the throttling of one or more burners on the tubes present in their vicinity, it is possible to quickly determine the appropriate burner map to be throttled in response to the throttling of different burners, thanks to a quick estimation of the TST of the tubes (point 3 iii) above).

[0072] Obtaining a map of the burner to be throttled is crucial. The solution according to the invention as disclosed below allows to significantly accelerate the realization of said map by proposing a simplified physical model of the influence of burner throttling on the tube surface temperature TST.

[0073] To develop rules to control the effect of burner throttling on tube surface temperature variations, the inventors first used numerical simulations to establish the effect of burner throttling on tube surface temperature. The simulations were performed using the in-house SMR3D solver, which is designed to calculate heat transfer between the combustion chamber and the tubes. More specifically, 3D computational fluid dynamics (CFD) was used for the combustion chamber, and a 1D model was used to account for the kinetics of the reforming reaction, using either known suitable solvers or in-house solvers. The simulations were performed in a representative domain of the furnace (a portion of the furnace representing the entirety) to define a simplified physical model.

[0074] Analysis of the tube surface temperature in response to various burner throttlings led the inventors to define the following behavioral rules (points 1 to 3 below and Figure 4a and Figure 4b 、 Figure 5a and Figure 5b 、 Figure 6a 、 Figure 6b and Figure 6c ):

[0075] Point 1. Figure 4a and Figure 4b The effect of throttling a single burner on the TST of the surrounding tubes is shown, and the effect of throttling the outer burner ( Figure 4a ) and throttling of the internal burner ( Figure 4b ) behave differently:

[0076] - For both the outer and inner burners, throttling the burner significantly affects the two nearest rows of tubes, hereinafter referred to as the "affected rows", with negligible effects on the more distant rows of tubes. For the outer burner, the two affected rows are on the same side of the throttled burner, while for the inner burner, the two affected rows are on each side of the throttled burner;

[0077] - outer burner throttling has a greater effect on the nearest row of tubes and a smaller effect on the second affected row, whereas inner burner throttling distributes the effect over the nearest row of tubes on each side;

[0078] - In each affected row, the throttling burner has the greatest effect on the nearest tube, - said tube being referred to hereinafter as the "affected tube", -, wherein this effect decreases rapidly when moving away from the throttling burner; the number of tubes affected by burner throttling is estimated to be 4×N 管 / N 燃烧器 (Obviously rounded to an integer), where N 管 is the number of tubes in a row and N 燃烧器 is the number of burners in a row.

[0079] Point 2. If Figure 5a and Figure 5b As shown, thanks to the numerical simulations, it is also advantageously found that throttling burners has a cumulative effect on the TST of the affected tubes. Figure 5a shows the results of a single simulation of two burner throttling, Figure 5b The results of adding two simulations of individual throttling are shown.

[0080] Point 3. Similarly, Figure 6a 、 Figure 6b and Figure 6c As shown, the TST variation is proportional to the power of the throttled burner. Due to simulations, it has been determined that the tube surface temperature variation (TST) and the power ratio are approximately proportional.

[0081] These behavioral rules for the evolution of the TST of tubes in an SMR furnace in response to burner throttling help to develop a simplified physical model to predict the effect of burner throttling on the tube surface temperature very instantaneously.

[0082] Then, to accomplish the present invention, the model is combined with an optimization algorithm to automatically obtain a burner map to be throttled. An internal optimization algorithm or an optimization algorithm known in the art, such as a black box optimization algorithm, can be used.

[0083] Based on the above, a method is established that aims to improve the distribution of heat transferred to the tubes, depending on parameter(s) and selected objectives, for example by reducing the maximum tube temperature or by alleviating tube or collector temperature differences.

[0084] The object of the present invention is therefore to propose a method for improving endothermic processes occurring in a furnace comprising tubes filled with a catalyst for the chemical conversion of a gaseous feed and positioned vertically in rows within the furnace, wherein the burners are installed in rows, wherein the inner burners of a row are placed between two rows of tubes and the outer burners of a row are placed between a row of tubes and a side wall parallel to the row of tubes, wherein the heat transfer from the burners to the tubes is correlated with the flow rate, wherein the method improves the temperature homogenization of the tubes by throttling a portion of the burners, characterized in that the method comprises the following steps:

[0085] step a) optionally acquiring information about the actual tube temperature of at least one tube closest to the burner in standard and throttle conditions,

[0086] step b) acquiring information about the actual tube temperatures of the tubes present in the furnace, all burners present in the furnace being under standard unthrottled conditions,

[0087] Step c) obtaining a burner map to be throttled, comprising:

[0088] c1) selecting at least one parameter representative of the performance of the furnace and a target for improvement,

[0089] c2) selecting at least one or more power ratios for burner throttling,

[0090] c3) using the information from step b) and a simplified physical model of the effect of burner throttling on the tube surface temperature of the furnace tubes to individually understand the effect of burner throttling on the target selected in step c1) and individually identify the burners to be throttled,

[0091] c4) obtaining a burner map of the furnace, wherein the burners to be throttled are individually identified,

[0092] Step d) throttling the burner according to the map obtained in step c4).

[0093] The method of the present invention may present one or more of the following variations, alone or in combination:

[0094] According to a preferred variant, the simplified physical model of the effect of burner throttling on the tube surface temperature is based on the following behavior rules 1) to 3):

[0095] 1) Throttling a burner significantly affects the two closest rows of tubes, which are on the same side for the outer burner and on each side of the burner for the inner burner; throttling an outer burner has a greater effect on the first row of tubes than on the second row, while throttling an inner burner distributes the effect to the closest rows of tubes on both sides; in any affected row, burner throttling has the greatest effect on the closest tubes, and the effect on tubes decreases rapidly as one moves away from the throttling burner; the number of tubes affected by burner throttling is estimated to be 4×N 管 / N 燃烧器 , for the affected tube, the tube surface temperature changes in proportion to the distance from the throttling burner

[0096] where N 管 = the number of tubes in a row and N 燃烧器 = number of burners in a row;

[0097] 2) The effect of throttling burners on tube temperature accumulation;

[0098] 3) The tube temperature change is proportional to the power ratio τ, where τ is:

[0099]

[0100] Among them, P 非节流 is the power of the burner in standard operating mode, and ΔP is the power change caused by throttling, ΔP<0 when the burner power decreases due to burner throttling and ΔP>0 when the burner power increases due to power redistribution among the non-throttling burners in the furnace.

[0101] Advantageously, the one or more parameters of step c1) are selected from the following: maximum tube surface temperature, tube surface temperature difference, synthesis gas collector temperature difference, and the improvement goal is to reduce the parameter value.

[0102] Preferably, in step c3), the simplified physical model is combined with an optimization algorithm, which is either an internal optimization algorithm or an optimization algorithm known in the art, such as a black-box optimization algorithm.

[0103] Throttling of the burner may be achieved by partially closing at least one valve mounted on at least one of the fuel or oxidant streams, preferably on the fuel stream, and more preferably on the secondary fuel stream.

[0104] Advantageously, the power ratio τ% of the throttling burners is between 90% and 50%, preferably between 80% and 60%.

[0105] Preferably, the information about the actual tube temperature in step b) is the tube surface temperature, preferably obtained by means of thermocouples or pyrometers or infrared cameras, or derived from the syngas collector temperature, preferably obtained by means of thermocouples.

[0106] The method may comprise a calibration step a), wherein actual tube temperature measurements are performed for one or more tubes affected by a throttled burner, preferably an external burner, at least for the tubes in front of said burner under non-throttled and throttled conditions.

[0107] The method of the invention allows debottlenecking a process / plant for producing hydrogen (or another gas or gas mixture) from synthesis gas obtained by steam gas reforming by homogenizing the temperature of the tubes by adjusting the power distribution.

[0108] To reduce the maximum TST below the MOT, the load on the unit can be increased until the maximum tube temperature reaches the MOT.

[0109] Another object of the invention is to propose a furnace comprising tubes filled with a catalyst for the chemical conversion of a gaseous feed and positioned vertically in rows inside the furnace, the burners being mounted in rows, wherein the inner burners of a row are placed between two rows of tubes and the outer burners of a row are placed between a row of tubes and a side wall parallel to the row of tubes, wherein the heat transferred from the burners to the tubes is correlated with the flow rate, characterized in that a portion of the burners is individually throttled according to a diagram obtained by any of the above methods.

[0110] The invention and its advantages will be described in more detail in the following examples on the basis of the accompanying drawings.

[0111] In the attached figure:

[0112] Figure 1A typical arrangement of tubes and burners is shown, using a 3D representation of a top-fired furnace for syngas synthesis;

[0113] Figure 2 shows a top view of a top-fired furnace, highlighting the tube and burner placement, and also highlighting representative domains;

[0114] Figure 3a The main flow entering the burner in the standard operating mode is shown;

[0115] Figure 3b Shows the entry in operating mode Figure 3a The main flow of the burner is throttled by partially closing the main fuel valve and the secondary fuel valve;

[0116] Figure 4a The effect of throttling the outer burner (power ratio τ of 50%) on the TST of the tube in a representative domain is shown;

[0117] Figure 4b The effect of throttling the inner burner (power ratio τ of 50%) on the TST of the tube in the same representative domain is shown;

[0118] Figure 5a The effect of throttling the outer and inner burners simultaneously on TST (power ratio τ is 50%) is shown. This is the result of a simulation with both burners throttled.

[0119] Figure 5b The effect of throttling both the outer and inner burners on TST (power ratio τ is 50%) is shown, as a sum of two separate simulations for one burner.

[0120] Figure 6a The effect of throttling the outer burner on TST is shown for a power ratio τ of 0% (valve closed);

[0121] Figure 6b The effect of throttling the outer burner on the tube surface temperature is shown for a power ratio τ of 75% (valve open 3 / 4);

[0122] Figure 6c The relationship between TST variation and power ratio τ is shown;

[0123] Figure 7a The method of the present invention is shown to be Figure 1 A first example of a burner map to be throttled obtained for a furnace of the type shown in FIG. 1 , wherein the two power ratios τ%=60% and τ%=80% are selected and the target parameters are TST maximum and TST difference;

[0124] Figure 7bThe experimental TST curves are shown: for the standard operating mode of the burners (ie all burners are in the standard non-throttling state) and for the Figure 8a Optimized burner throttling of the diagram;

[0125] Figure 8a A second example of a burner diagram to be throttled obtained by the method of the invention for the same furnace is shown, applying a single power ratio τ%=60% chosen, with the aim of reducing the parameters TSTmax and Syngas Collector Temperature Difference.

[0126] Figure 8b Shows the standard operating mode for the burner and the Figure 9 The experimental TST curve of the optimized burner throttling shown in FIG;

[0127] Figure 9 The evolution of the syngas collector temperature gap for the standard configuration and the Figure 7a The first example and Figure 8a The second example scene implements the throttling diagram.

[0128] The following detailed description of the accompanying drawings and examples will facilitate understanding of the present invention.

[0129] Figure 1 is a 3D perspective view of a furnace; more precisely, it shows a typical layout of a top-fired furnace 1 for producing synthesis gas. Tubes 2 containing reforming catalyst are arranged in rows within the top-fired furnace 1. A gaseous mixture of feed gas, methane, and steam is supplied to the inlet of tubes 2 at the top of the furnace. As it flows through the catalytic bed toward the bottom of tubes 2, the process gas is converted and discharged as synthesis gas: a mixture of primarily hydrogen and carbon monoxide (also known as syngas). Burners 3 are placed between the rows of tubes, so that each row of tubes consists of two rows of burners; the orientation of the rows of tubes and burners is also along the X-axis. The flue gases generated by the combustion of the fuel and air mixture in the burners 3 are discharged through exhaust ducts 4 installed at the bottom of the furnace, parallel to the rows. The direction perpendicular to the rows is designated as the Y-axis.

[0130] Figure 2A top-down view of a top-fired furnace 1 is presented. The furnace comprises eight rows of five tubes (50 tubes 2 per row) and nine rows of burners 6a and 6b, each row containing 15 outer burners 3a and inner burners 3b, parallel to the rows of tubes. The burners are arranged into two outer rows 6a and seven inner rows 6b. Each outer burner 3a in each row 6a extends between one of the two side walls 7 and a row of tubes, while each inner burner 3b in each inner row 6b (referred to as an inner burner) is surrounded by a row of tubes on each side. This organization allows the outer burners 3a, heating one row of tubes, to operate at a lower power level than the inner burners, heating two rows of tubes. The figure highlights in gray a representative domain 8 (within the meaning of the present invention) comprising a subset of four partial rows of tubes (17 tubes per row), heated by three partial rows of inner burners and one partial row of half-inner burners (these burners are cut in the middle along a symmetry plane S, which is parallel to the X axis and parallel to the plane W representing the side wall). Along the Y axis, the representative domain 8 is also bounded by two symmetry planes perpendicular to the symmetry plane S. This representative domain 8 is used to illustrate the present invention, for simulations, and also for figures related to the simulations.

[0131] Figure 3a The main streams entering the burner 3 are shown, each stream being carried by a conduit equipped with a valve; that is: oxidant stream 9 flows through valve 10, natural gas (primary fuel) stream 11 flows through valve 12, and tail gas (secondary fuel) stream 13 flows through valve 14. The furnace is operated in standard mode with the valves fully open and the flow of the streams unrestricted.

[0132] As mentioned above, the present invention aims to reduce the temperature of the hottest tubes; to achieve this goal, the invention aims to reduce the power of some specifically selected burners by reducing, preferably, the flow rate of one or more fuel streams. As a result, the heat transferred to the affected tubes is reduced, resulting in a reduction in their temperature.

[0133] Figure 3b Streams 9, 11, and 13 entering a burner operating according to the present invention are shown. In a preferred mode of the invention, the power emanating from exhaust gas stream 13 (the primary fuel contributor under standard conditions) and natural gas stream 11 is reduced by throttling valves 14 and 12, respectively. This reduction in burner power provided by throttling the fuel valves will correspondingly result in a reduction in the TST of the adjacent tubes, according to the aforementioned rules of conduct 1 through 3. Note that a reduction in burner power can also be achieved by throttling only one of the fuel valves.

[0134] Figure 4a Shows that, for Figure 2The effect of throttling the external burner 3a on the TST of the different tubes is shown in the representative domain 8 defined in [1]. The TST is obtained using a solver (known per se) that calculates the heat transfer between the furnace's combustion chamber and the tubes. In the presented case, the burner is throttled to achieve a 50% power ratio.

[0135] Figure 4b The effect of throttling the inner burner 3b on the TST of different tubes in the same domain is shown. The TST is obtained using the same solver. The power of the throttled inner burner is also reduced to achieve a 50% power ratio.

[0136] In two Figure 4a and Figure 4b Above, for each tube in a representative field, the change in TST (ΔTST, also referred to as ΔT in the figure) is presented by reference to the temperature measured without throttling (i.e., under standard conditions); ΔTST varies from "0°C" for the tube not affected by throttling to "-13°C" for the most affected tube. ΔTST is represented using a grayscale that varies from white to black, from white for the unaffected tube (ΔT=0°C) to black for the most affected tube (ΔT=-13°C).

[0137] therefore, Figure 4a and Figure 4b Given the above behavior rule 1, the behavior rule can be summarized as follows:

[0138] - Throttling the burner will affect the two closest rows of tubes; the effect on the rows of tubes is negligible relative to the throttling burner; however, it should be noted that for the outer tubes, the two closest rows are on the same side, and for the inner tubes, the two closest rows are on each side of the throttling burner;

[0139] -Throttling of an outer burner has a greater impact on its nearest tube bank than that of an inner burner;

[0140] - Within each affected row, the throttling burner will have the greatest effect on the nearest tube; the effect on the tubes decreases rapidly as one moves away from the throttling burner. The number of tubes affected by burner throttling is estimated to be 4×N 管 / N 燃烧器 (rounded to an integer), where N 管 is the number of tubes in a row and N 燃烧器 is the number of burners in a burner row.

[0141] Figure 5aThe effect of throttling two burners simultaneously is shown, more precisely the result of a single simulation that takes into account the fact that two burners are throttled: the outer burner 3a and the inner burner 3b. For both burners, the burner power is reduced by 50%.

[0142] Figure 5b The sum of the two calculated effects of individual throttling for identical burners with identical power ratios is shown.

[0143] Figure 5a and Figure 5b The comparison demonstrates the aforementioned behavior rule 2: throttled burners have a cumulative effect on the tube surface temperature. This means that the effect of a group of throttled burners on a given TST is simply the sum of the individual contributions.

[0144] Figure 6a and Figure 6b The effect of different power ratios for the outer burner 3a on the TST of the most affected tube is shown. Figure 6a In the example, the power ratio of the burner (black square) is 0% (burner off). Figure 6b In FIG, the power ratio of the burner (also a black square) is 75%, which means that the burner delivers 75% of its nominal power.

[0145] The change in tube surface temperature (called ΔT in the figure) is presented with reference to the temperature measured without throttling, ranging from 0°C for the tubes not affected by throttling to -23°C for the most affected tubes. The decrease in TST temperature for each tube is represented using a gray scale that varies from white to black, with ΔT being white for 0°C to -23°C for Figure 6a The black plot shows the change in ΔT of -23°C on the most affected tube. These plots demonstrate the fact that the tube surface temperature decreases in proportion to the change in power ratio. The closer the power ratio is to zero, the greater the decrease in tube surface temperature TST.

[0146] Figure 6c Use from Figure 4a 、 Figure 6a and Figure 6b The information in FIG. 1 shows the variation of TST as a function of τ% for the tube closest to the throttled burner. The black squares ■ represent three cases where the power ratio is reduced (τ% = 0%, τ% = 25%, and τ% = 75%), while the empty squares □ represent the case where the burner power remains unchanged (τ% = 100%), and the "diamond ◇" represents the case where the burner represented is not throttled while a large number of other burners in the furnace are throttled, causing the fuel flow to be redistributed to the non-throttled burners in the furnace and thus their power to increase, thereby making the power ratio greater than 1. This figure proves the above-mentioned behavioral rule 3: the tube surface temperature variation is proportional to the power ratio, as shown by the solid line for the tube closest to the throttled burner.

[0147] By using the ratio (Rule 3 of the present invention) - a rule that is effective in the affected area around the burner - the variation of the TST for any tube affected by throttling can be obtained, as shown by the dotted lines for tubes further from the throttling burner.

[0148] This is because the inventors have highlighted these rules governing the evolution of the TST of tubes in an SMR furnace in response to burner throttling (i.e. the rapid decrease of the tube surface temperature with respect to the distance from the throttled burner), the linear influence of the power ratio on the tube surface temperature, the cumulative effect of throttling several burners - they have been able to develop the method of the present invention to determine the optimization map for burner throttling.

[0149] When the method of the invention is used for the first time in a steam reformer, or in order to adapt to different operating conditions or to take into account significant variations, the simplified physical model can be calibrated. The calibration step can include performing actual tube temperature measurements for one or more tubes affected by the throttling burner, at least for the tubes in front of the burner under non-throttling conditions and under throttling conditions; the burner is preferably an external burner, and additional measurements can be performed for the most distant tube affected by the burner throttling. The actual TST variation is obtained by comparing the two values ​​under non-throttling conditions and under throttling conditions, which will give Figure 6c The slope of the median line(s).

[0150] The input data for step b) is not limited to the tube surface temperature; it can also be the syngas collector temperature; in this case, the tube temperature is derived from the corresponding syngas collector temperature. Typically, this derivation assumes that the tubes in a row are at the same temperature.

[0151] The proposed method is used to calculate the following tools:

[0152] -Optimized algorithms,

[0153] - a simplified physical model according to the invention,

[0154] - one or several parameters and their associated targets to be achieved, which will lead to an improvement in process performance - parameters and associated targets are for example: TST gap and its reduction, and / or maximum TST and its reduction, and / or syngas collector temperature gap and its reduction. These parameters are currently preferred for practical reasons: good representation of the furnace behavior, easy collection of input data, however, the invention is not limited to this choice of input data and parameters, other parameters and targets may also be chosen,

[0155] - One or more power ratios to apply

[0156] Example

[0157] The solution of the invention has been successfully implemented on site at a hydrogen production plant with a top-fired steam methane reformer furnace equipped with 400 tubes (organized in 8 rows of 50 tubes each) and 135 burners (formed in 9 rows of 15 burners each).

[0158] Thanks to the invention, by applying the taught method, it is possible to automatically and quickly determine the favorable map(s) of the burner to be throttled relative to given target(s).

[0159] First, a single set of TST measurements was performed under standard (i.e., non-throttling) conditions; the measurements were performed from the first peephole level of the furnace using a pyrometer;

[0160] The simplified physical model of the invention based on the above-mentioned behavior rules 1 to 3 is used to obtain a suitable burner diagram to be throttled,

[0161] Due to its fast computational execution time, the combination of the simplified physical model and the internal optimization algorithm of the present invention allows the automatic and rapid deriving of the burner diagram to be throttled from the initial TST measurements obtained under standard conditions in order to improve the process as expected (i.e., to achieve the desired target).

[0162] The timeliness with which the burner map to be throttled is achieved when applying the solution of the invention allows comparing the performance of several burner maps obtained for different power ratios and / or different parameters and their targets.

[0163] Get the value for Example 1 Figure 7a and for Example 2 Figure 8a The burner diagram presented takes approximately 10 minutes to throttle. Both experiments were performed on the same reformer under the same standard conditions.

[0164] For the implementation of the on-site throttling method, the reduction of burner power is achieved by appropriately reducing the exhaust gas flow rate due to the valve on the exhaust gas flow.

[0165] The following is about Example 1 Figure 7a and Figure 7b And about Example 2 Figure 8a and Figure 8b Detailed methods applied to these two examples are presented.

[0166] Example 1.

[0167] Figure 7a The burner diagram to be throttled obtained by the method of the invention is shown. In this first example, two power ratios are selected: 80% and 60%.

[0168] The following target parameters were selected: reduction of the maximum TST (ie the TST of the hottest tube) and reduction of the TST gap; an additional target was to minimize the number of burners to be throttled.

[0169] In the diagram of burners to be throttled presented in the figure, burners identified as requiring an 80% power ratio are represented as black diamonds "◆", burners identified as requiring a 60% power ratio are represented as black squares "■", and burners without restrictions, i.e., burners with unchanged fuel flow (except for a small redistribution part), are represented as empty squares "□".

[0170] Figure 7a It is shown that in order to achieve the goal of reducing the maximum TST and reducing the TST gap while minimizing the number of burners to be throttled, 24 burners (individually identified) need to be throttled, 13 burners need to be throttled to a power ratio of 80% and 11 burners need to be throttled to a power ratio of 60%.

[0171] Figure 7b The TST curves obtained from the same example are shown: the experimental TST curve measured in the standard operating mode of the burner is represented by the gray line; Figure 8a , (shown as black circles "●"), the TST curve was measured after the plant was stabilized with all burners throttled. The MOT (maximum operating temperature) is represented as a dotted line.

[0172] Under standard conditions, such as Figure 7b As shown, the TST of some tubes exceeds the MOT (indicated as the dashed line at 949° C.) They are too hot, and this may lead to tube failure.

[0173] Due to the implementation of the method of the present invention, by Figure 7a The diagram shown throttles the burner and all TSTs are maintained below the MOT limits while the TST gap is reduced by 15°C; this is considered a good enough solution.

[0174] Example 2.

[0175] Figure 8a Another burner diagram to be throttled obtained by applying the method of the present invention to the same furnace is shown. In this example, it is decided that:

[0176] - Apply a single power ratio of 60%, and

[0177] The following target parameters were selected: reduction of the maximum TST and reduction of the syngas collector temperature gap. In the figure, burners requiring a 60% throttling ratio are represented by "black squares ■"; unthrottled burners, i.e., burners with a constant fuel flow (except for a small redistribution), are represented by "empty squares □."

[0178] Figure 8a Thirty-seven burners (identified individually) are shown which, when throttled, allow the target to be achieved.

[0179] Figure 8b Shows when applied Figure 8a TST curve of the tube obtained when the solution is: the experimental tube temperature curve measured in the standard operating mode of the burner (as a continuous gray line) and the Figure 8a The TST temperature curve of the burner diagram is shown as a "black circle ●". The MOT (maximum operating temperature) is represented as a dotted line.

[0180] like Figure 8b As shown, under standard conditions, the TST of several tubes exceeded the maximum operating temperature.

[0181] Due to the implementation of the method of the present invention, by Figure 8a The burner is throttled to maintain the TST of all tubes below the MOT, while reducing the temperature difference of the syngas in the collector (as shown in FIG. Figure 9 This is considered a good enough solution for the operation of the device.

[0182] at last, Figure 9 Presented the on-site realization Figure 7a and Figure 8a Evolution of the measured syngas collector temperature difference when throttling diagram is shown.

[0183] -According to Figure 8a The burners were throttled on site (Example 1 has 24 throttling burners), and the temperature difference of the syngas collector was reduced by 15°C;

[0184] -Through on-site implementation Figure 8a The throttling diagram with 37 throttling burners (Example 2) reduces the syngas collector temperature difference by 25°C.

[0185] Thanks to the invention, the time required to establish a burner map to be throttled is approximately ten minutes, which is much shorter than that required by previously known solutions; this is because the simplified physical model identified and applied by the inventors significantly reduces the duration of determining the burner map to be throttled.

[0186] - This makes the solution a simple remedy that allows alleviating the uneven heat transfer between the reformer tubes;

[0187] - The burner is easily throttled according to the throttling burner diagram; thanks to the valves installed on the pipeline, the operator can manually adjust the burner power according to the diagram, or transmit the information to the "device control" and automatically throttle the burner according to the diagram in a single operation;

[0188] -The solution can be used frequently on-site to facilitate operators’ decision-making in their daily work;

[0189] - This solution allows to build several burner maps with different throttling power ratios in order to select the most suitable solution.

Claims

1. A method for improving an endothermic process occurring in a furnace comprising tubes filled with a catalyst for the chemical conversion of a gaseous feed and positioned vertically in rows within the furnace, wherein burners are fed with a primary fuel flow and a secondary fuel flow and an oxidant flow and are mounted in rows, wherein the inner burners of a row are placed between two rows of tubes and the outer burners of a row are placed between a row of tubes and a side wall parallel to the row of tubes, wherein the heat transfer from the burners to the tubes is correlated with the flow rate, wherein the method improves the temperature homogenization of the tubes by throttling a portion of the burners, characterized in that The method comprises the following steps: step b) obtaining information about the actual tube temperatures of the tubes present in the furnace, all burners present in the furnace being under standard non-throttled conditions, step c) obtaining a map of burners to be throttled, comprising: c1) selecting at least one parameter representative of the performance of the furnace and a target for improvement, c2) selecting at least one or more power ratios for burner throttling, c3) utilizing the information from step b) and a simplified physical model of the effect of burner throttling on the tube surface temperature of tubes in the furnace to individually understand the effect of burner throttling on the target selected in step c1) and individually identify burners to be throttled, c4) obtaining a burner map of the furnace, wherein the burners to be throttled are individually identified, and step d) throttling the burners according to the map obtained in step c4); The simplified physical model of the effect of burner throttling on the tube surface temperature is based on the following behavioral rules 1) to 3): 1) Throttling a burner significantly affects the two closest rows of tubes, which are on the same side for the outer burner and on each side of the inner burner for the inner burner; throttling an outer burner has a greater effect on the first closest row of tubes than on the second row, while throttling an inner burner distributes the effect to the closest rows of tubes on both sides; in any affected row, burner throttling has the greatest effect on the closest tubes, with the effect on these tubes decreasing rapidly as one moves away from these throttled burners; the number of tubes affected by burner throttling is estimated to be 4×N 管 / N 燃烧器 For the affected tubes, the tube surface temperature changes in proportion to the distance from the throttling burner. where N 管 = the number of tubes in a row and N 燃烧器 = number of burners in a row; 2) The effect of the throttling burner on the temperature accumulation of the tube; 3) The temperature change of these tubes is proportional to the power ratio τ, where τ is: Among them, P 非节流 is the power of the burner in standard operating mode, and ΔP is the power change caused by throttling, ΔP<0 when the burner power decreases due to throttling of the burner and ΔP>0 when the burner power increases due to power redistribution among the non-throttling burners in the furnace.

2. The method according to claim 1, characterized in that The method comprises the following steps: step a) calibrating the simplified physical model of step c3) by measuring one or more actual tube temperatures of at least one tube affected by burner throttling in standard and throttled conditions.

3. The method according to claim 1, characterized in that The one or more parameters in step c1) are selected from the following: maximum tube surface temperature, tube surface temperature difference, synthesis gas collector temperature difference, and the goal of improvement is to reduce the parameter value.

4. The method according to claim 2, characterized in that The one or more parameters in step c1) are selected from the following: maximum tube surface temperature, tube surface temperature difference, synthesis gas collector temperature difference, and the goal of improvement is to reduce the parameter value.

5. The method according to claim 1, wherein In step c3), the simplified physical model is combined with an optimization algorithm, which is an optimization algorithm known in the art.

6. The method according to claim 2, characterized in that In step c3), the simplified physical model is combined with an optimization algorithm, which is an optimization algorithm known in the art.

7. The method according to claim 3, characterized in that In step c3), the simplified physical model is combined with an optimization algorithm, which is an optimization algorithm known in the art.

8. The method according to claim 4, characterized in that In step c3), the simplified physical model is combined with an optimization algorithm, which is an optimization algorithm known in the art.

9. The method according to claim 5, characterized in that The optimization algorithm known in the art is a black box optimization algorithm.

10. The method according to claim 6, characterized in that The optimization algorithm known in the art is a black box optimization algorithm.

11. The method according to claim 7, characterized in that The optimization algorithm known in the art is a black box optimization algorithm.

12. The method according to claim 8, characterized in that The optimization algorithm known in the art is a black box optimization algorithm.

13. The method according to any one of claims 1 to 12, characterized in that The throttling of these burners is obtained by partially closing at least one valve installed on at least one of the fuel flow or the oxidant flow.

14. The method according to claim 13, characterized in that The throttling of these burners is obtained by partially closing at least one valve installed on the fuel flow.

15. The method according to claim 14, characterized in that The throttling of these burners is obtained by partially closing at least one valve installed on the secondary fuel flow.

16. The method according to any one of claims 1 to 12, characterized in that The power ratio τ% of the throttling burner is between 90% and 50%.

17. The method according to claim 13, wherein The power ratio τ% of the throttling burner is between 90% and 50%.

18. The method according to claim 14, characterized in that The power ratio τ% of the throttling burner is between 90% and 50%.

19. The method according to claim 15, characterized in that The power ratio τ% of the throttling burner is between 90% and 50%.

20. The method according to claim 16, wherein The power ratio τ% of the throttling burner is between 80% and 60%.

21. The method according to any one of claims 1 to 12, characterized in that The information about the actual tube temperature in step b) is the tube surface temperature.

22. The method according to claim 13, wherein The information about the actual tube temperature in step b) is the tube surface temperature.

23. The method according to claim 14, wherein The information about the actual tube temperature in step b) is the tube surface temperature.

24. The method according to claim 16, wherein The information about the actual tube temperature in step b) is the tube surface temperature.

25. The method according to claim 17, wherein The information about the actual tube temperature in step b) is the tube surface temperature.

26. The method according to claim 20, wherein The information about the actual tube temperature in step b) is the tube surface temperature.

27. The method according to claim 21, characterized in that The tube surface temperature is obtained with the aid of a thermocouple or a pyrometer or an infrared camera, or is inferred from the syngas collector temperature.

28. The method according to claim 27, characterized in that The syngas collector temperature was obtained with the aid of a thermocouple.

29. The method according to any one of claims 1 to 12, characterized in that The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

30. The method according to claim 13, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

31. The method according to claim 14, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

32. The method according to claim 16, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

33. The method according to claim 17, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

34. The method according to claim 20, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

35. The method according to claim 21, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

36. The method according to claim 22, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

37. The method according to claim 28, wherein The method comprises a calibration step a) in which actual tube temperature measurements are performed for one or more tubes affected by a throttling burner, at least for the tubes in front of said burner under non-throttling and throttling conditions.

38. The method of claim 29, wherein the burner is an external burner.

39. A furnace comprising reforming tubes filled with a catalyst for the chemical conversion of a gaseous feed and positioned vertically in rows within the furnace, the burners being mounted in rows, wherein the inner burners of a row are placed between two rows of tubes and the outer burners of a row are placed between a row of tubes and a side wall parallel to the rows of tubes, wherein the heat transfer from the burners to the tubes is correlated with the flow rate, characterized in that A portion of these burners is throttled individually according to a map obtained by the method according to any one of claims 1 to 38 .

Images