APPARATUS AND METHOD FOR INVESTIGATING NAPHTHA REFORMATION PROCESSES

Description

[0001] The present invention relates to a method for investigating naphtha reforming processes in catalyst test apparatus with parallel reactors.

[0002] Naphtha reforming is used for the catalytic conversion of naphtha and light oils into higher-value fuels (gasoline) and chemicals (aromatics). Due to the high demand for fuels and basic chemicals, naphtha reforming is of great economic and technical importance. Furthermore, naphtha reforming produces hydrogen, which is urgently needed for hydroprocessing in refineries. The further development and improvement of naphtha reforming is of interest, particularly to increase the yield of reformate (gasoline, aromatics) per barrel of naphtha and thus conserve valuable resources.

[0003] This paper concerns the use of catalyst test equipment for carrying out naphtha reforming in the laboratory or in pilot plants, in order to improve the process parameters for naphtha reforming and / or to develop new catalysts. The test results obtained in pilot plants enable improvements to large-scale processes and are of great value to the catalyst manufacturing and petrochemical industries, where naphtha processing takes place.

[0004] Even minor improvements in large-scale naphtha reforming have a significant impact on its economic viability and associated energy costs, given the substantial industrial scale. However, laboratory methods using catalyst test equipment have limitations, as replicating large-scale processes in laboratory settings is subject to numerous sources of error and fluctuations. Consequently, even minor improvements may be missed or only partially detected due to variations in measurement accuracy.

[0005] A particularly important parameter that must be monitored and / or optimized during naphtha reforming is the so-called "octane number." The octane number indicates the knock resistance of gasoline and thus serves as a measure of gasoline quality. The knock resistance of gasoline is determined on engine test benches, and a distinction is made between "RON" (i.e., the Research Octane Number) and the MON (Motor Octane Number). The MON value refers to the knock resistance of gasoline measured at an engine speed of 900 rpm. The RON value refers to the knock resistance of gasoline measured at an engine speed of 600 rpm. Half a liter of gasoline mixture is required for both RON and MON determination on an engine test bench. In the USA, these values ​​refer to the average of the values ​​determined for RON and MON (i.e., the average of the RON and MON values).The octane rating is calculated as (RON+MON) / 2).

[0006] Determining the octane rating in laboratory equipment using analytical characterization methods other than those used to measure the knock resistance of gasoline on engine test benches during reforming is of great importance. Firstly, it avoids the costly and time-consuming preparation of half-liter samples. Secondly, this would provide insufficient temporal resolution of the octane rating, as these samples would have to be collected over a long period and would therefore only yield an average value over that time interval.

[0007] It is generally known in the art how the octane number can be calculated using analytical measurements, particularly by gas chromatography and near-infrared spectroscopy. To determine the octane number using gas chromatography, the composition of the reformate is first determined, and then the octane number is calculated by applying octane number models. For an overview, reference is made to the publication by Knop et al. in Fuel 115 (2014) 666-673, in which various octane number models are presented and evaluated.

[0008] Methods for investigating naphtha reforming in the laboratory are known in principle. In naphtha reforming, naphtha is brought into contact with mostly platinum-containing bifunctional catalysts, in which finely dispersed platinum is suspended on (usually chlorinated) aluminum oxide. In addition, bimetallic or multimetallic catalysts are increasingly being used.

[0009] Catalyst test setups for the laboratory investigation of catalysts in general are known in principle. For example, US patent US 7,118,917 B2 by Bergh et al. discloses a method and apparatus with four or more parallel reactors, in which individual heating of the reactors is possible. Furthermore, the reactors can be heated to different temperatures along the reactor axis using different heating zones, thereby ensuring more precise temperature control within the reactors. Reaction apparatuses with parallel reactors, which are also suitable in principle for testing catalysts, are known. An example of this is US patent 2002 / 0182735 A1 by Chuck Kibby et al., filed with the US Patent Office on May 28, 2002, by Chevron Texaco Corporation.In parallel reactors, the products can be examined with regard to their properties, with the aim of obtaining optimized catalysts for a predetermined product stream using combinatorial methods.

[0010] Publications cited in the search report: WO-A 02 / 14854 (Chuck Kibby et al.) discloses a method for investigating catalysts using reactors with microchannels. The reactors are in the form of laminate layers. The advantage is that a large number of parallel reaction channels can be arranged in a single assembly.

[0011] In his publication (Oil and Gas Journal, 492, Vol. 77, No. 39, pp. 191-200), K.P. Kennedy describes a software control system for supporting catalytic reforming, which is used in large-scale production plants. Numerous operating parameters influence product formation and the associated octane rating.

[0012] In the publication by H. l'Oncle (L'Industrie du Petrole en Europe Gaz-chmie, 36, 388 (1968), p. 54-55), the automation of a pilot plant for carrying out catalytic reforming is described, whereby the composition of the product stream is analyzed and the control parameters are adjusted using an algorithm. The octane rating is also determined in this process.

[0013] US patent 2013 / 0323853 discloses a method for monitoring large-scale reforming plants using a detailed analysis of hydrocarbon composition. Different GC methods for analyzing the octane number are described and compared.

[0014] One of the objectives of the present invention is to adapt and improve upon prior art methods for investigating catalysts in parallel plants, particularly with regard to naphtha reforming processes. A fundamental technical problem to be solved is the lack of accuracy of laboratory methods, which are subject to statistical errors, thereby severely limiting their significance or even rendering them insignificant on a laboratory scale.

[0015] The tasks mentioned here, and numerous others not mentioned here, are solved by an apparatus and a method that are described in more detail below: Apparatus

[0016] The apparatus according to claim 1 comprises a plurality of reactors arranged in parallel, preferably tubular reactors, wherein the apparatus is suitable for investigating naphtha reforming, wherein the apparatus further comprises a process control, a feed fluid supply (common to all or a plurality of the reactors), and each individual reactor has its own output line for product stream, as well as at least one online analysis unit, wherein each individual reactor of the plurality of reactors is further equipped with a separate heating system, wherein the apparatus is characterized in that the analysis unit is operatively connected to each output line for product fluid stream.

[0017] The analysis unit operates "online," meaning that there is at least a temporary fluidic connection between a reactor outlet and the sample inlet of the analyzer. This further means that the product fluid stream, or an aliquot thereof, is transferred directly into an analyzer; preferably, no manual manipulation of the sample is required at this point. The analysis unit can evaluate the analysis results directly and immediately and transmit them to the control unit. With such an "online" analysis unit, the time between the physical measurement of the parameter in the product fluid stream and the transfer of the measurement data to the control unit is preferably less than 100 milliseconds, more preferably less than 10 milliseconds, and even more preferably less than 1 millisecond. Overall, the times for transferring the measurement data to the control unit are very short.

[0018] In another embodiment, a multiport or selection valve is located in the respective output line of two or more reactors and on the input side of the analysis unit. This valve is designed such that an active connection is always established only between exactly one reactor output and the analysis unit at any given time (this corresponds to multiplexing).

[0019] The selection of the active component, and thus the reactor outflow, is controlled and initiated by the process control system. The results of the analysis unit are stored in the process control system in such a way that they can be assigned, both temporally and numerically, to the reactor from which the product fluid was taken. Changes to the variable parameters of a reactor are made based on the data assigned to it (= demultiplexing).

[0020] This online analysis unit allows the fluid flows from the individual output lines to be analyzed sequentially and / or in parallel.

[0021] The present invention relates to an apparatus for investigating naphtha reforming, wherein the apparatus comprises: a plurality of parallel reactors with reaction chambers (R1, R2,...), the individual reactors have a catalyst capacity in the range of 0.5 - 200 cc, a common reactant fluid supply for two or more of these reactors, a process control, an analysis unit wherein each individual reactor is equipped with an individually controllable heating device and has an output line for product fluid stream, characterized in that the analysis unit is operatively connected to each of these output lines for product stream, and that the analysis unit is functionally related to the control of the apparatus, wherein the analysis unit is suitable for the analysis of a gaseous product fluid stream, and the analysis unit together with the process control allows the determination and optimization of an octane number in the parallel arranged reactors, wherein the analysis unit is a gas chromatograph or the characterization of the gaseous product fluid stream is carried out by means of IR spectroscopy, and wherein the temperature of the respective reactor from whose output line the product fluid stream was analyzed can be adjusted. Proceedings

[0022] The problem according to the invention was further solved by means of an analysis unit, preferably an analysis unit comprising gas chromatography, in which the product composition for each reactor in the parallel plant is individually analyzed, wherein, in particular and preferably, the product fluid does not need to be condensed for this analysis, or at least not to a significant extent. Of particular importance for the method according to the invention is that the transport of the portion of the product fluid to be analyzed takes place in the gas phase and no condensation occurs.

[0023] The invention relates to a process according to claim 7, which is carried out using the apparatus according to the invention, and in which a naphtha-containing reactant fluid stream is supplied to each individual reactor of the plurality of reactors R1, R2, and is brought into contact with a catalyst in the respective reactor at a temperature in the range of 100°C to 600°C, preferably in the range of 400°C to 550°C, and is converted to a product fluid stream. Each individual product fluid stream is directed via the individual output lines to a product fluid intake or afterburner. The individual product fluid streams or partial streams of the individual output lines are each subjected to the following process steps: (i) analytical characterization using an analysis unit including the individual product fluid streams, wherein the analysis relates to the quantification of one or more components, preferably one or more groups of 10 to 20 components each, more preferably 10 groups of 1 to 100 components each, (ii) calculation of the octane number and comparison of the calculated octane number with a target value, (iii) calculation of the temperature difference necessary to maintain a constant octane number, (iv) adjustment of the temperature of the respective heating system surrounding the reactor from whose outlet line a product stream was analyzed. at least a repetition of process steps (i) to (iv) by analyzing the product fluid flow exiting the respective output lines and calculating the respective octane number that characterizes the product fluid flow.

[0024] According to a preferred embodiment, the present invention relates to a method for investigating naphtha reforming using an apparatus as described herein, wherein the naphtha-containing reactant fluid stream contains the following components: at least one hydrocarbon from the group with 5 to 14 carbon atoms, preferably a mixture of hydrocarbons with 5 to 14 carbon atoms, and hydrogen with a hydrogen content in the range of 0.5 to 60 vol.%, preferably 10 to 30 vol.%, which is supplied to the individual reactors R1, R2, and is brought into contact with a catalyst in the respective reactors at a temperature in the range of 100°C to 600°C, preferably in the range of 400°C to 550°C. and is converted into a product fluid stream, and this product fluid stream is subsequently fed to an output line, the process being characterized by the following process steps: (i) an analytical characterization of the individual product fluid streams using an analysis unit, wherein the analysis relates to the quantification of one or more components, preferably one or more groups of 1 to 300 components each, more preferably four groups of 20 to 150 components each, (ii) comparison of the result value of the quantification with a target value, wherein the comparison includes the input of an evaluation model, (iii) adjustment of a process parameter relating to the reaction process of the respective reactor from whose output line the product fluid stream was analyzed, a repetition of process steps (i) to (iii) by analysis of the product fluid stream,which exits from the respective output lines.

[0025] With regard to the term "common reactant fluid feed," reference is made to the publication of PCT application WO 2005 / 063372 A2 (with a priority date of December 23, 2003) by the same applicant. PCT application WO 2005 / 063372 A2 also discloses the meaning of the terms "pressure build-up gas" and "pressure maintenance gas." Furthermore, while the present invention is preferred for parallel reactor apparatus, it is not limited to parallel reactor apparatus but also includes single reactors constructed according to the present disclosure (without, however, having two or more reactors). In particular, the process according to the invention using a GC and without condensation also exhibits the advantages of the invention for a single reactor.

[0026] According to the present invention, the product fluid stream preferably contains less than 1% of condensed product fluid, and more preferably less than 0.5%.

[0027] The calculation of the octane number, as required for step (ii), is carried out using suitable octane number models derived from the product composition. A software-based algorithm performs, among other things, the evaluation of the chromatograms, determines the product composition, and calculates the octane number. The temperature adjustment is then calculated based on the corresponding, previously determined octane number-temperature relationship of the respective catalyst. This temperature difference is then transmitted to the respective reactor heaters to maintain a constant octane number.

[0028] The method according to the invention comprises process steps (i) to (iii), which form a control loop. The process steps are carried out and the control loop is executed for the entire duration of the method according to the invention. Preferably, the control loop is executed for each individual reactor for at least 2 to 10 iterations. The number of iterations of the control loop depends, among other things, on the total duration of the catalytic investigation. If the catalysts are tested for a period of weeks or months, the number of iterations of the control loop can be correspondingly high. The deactivation rate of the individual catalysts must also be taken into account with regard to the number of iterations. In the case of rapidly deactivating catalysts, it is preferred to run the control loop more frequently than in the case of slowly deactivating catalysts.

[0029] Preferably, chromatograms obtained using gas chromatography are evaluated.

[0030] The evaluation of the chromatograms preferably includes the identification of individual components or groups of components. In particular, the use of GC as an analytical method, especially in "online" operation, allows for high accuracy, which is a particular advantage for the method according to the invention. Preferably, the evaluation of the measurement data relies on reference data, which were preferably generated during calibration.

[0031] Due to the high complexity of multicomponent mixtures, online analytical characterization can take several minutes. This timeframe for analysis and the use of the preferred evaluation model has the advantage that the accuracy of the method according to the invention is significantly increased compared to prior art methods, particularly with regard to temporal resolution. The temporal resolution for investigating deactivation processes is also determined by the time required for the online analytical characterization.

[0032] The analytical characterization of the product fluid stream, preferably in online mode, is therefore the time-determining step. The online analytical characterization takes place in a time range of 0.1 to 300 minutes, preferably in a range of 1 to 150 minutes, and more preferably in a range of 10 to 40 minutes. In an alternative embodiment of the method according to the invention, the product fluid stream is characterized by IR spectroscopy. Just as with GC characterization, corresponding calibration data and methods are stored in the control system, which are used to analyze the components and determine the octane number.

[0033] This duration of the actual analysis must be distinguished from the duration of the data transfer from the analysis unit to the control unit, which was described above and is in the millisecond range. Therefore, the data transfer time is not the step that decisively influences the speed of the method according to the invention.

[0034] It should also be noted that many of the catalysts used for reforming naphtha-containing feed streams generally have sufficiently long lifetimes and do not deactivate very quickly. Therefore, the time required for performing the online analytical characterization and the intervals for repeating the control loop are adequate in relation to the deactivation rate of the individual catalysts.

[0035] Furthermore, in another embodiment, the experimental setup and the method can be configured such that the duration for repeating the individual iterations of the control loop can be adjusted accordingly. In particularly preferred embodiments of the method, this adjustment can, for example, consist of preferably using a multi-channel GC. The multi-channel GC allows the product fluid flows from several reactors to be analyzed simultaneously. It is also preferred to equip the apparatus with more than one online analysis unit. Particularly preferred embodiments include an apparatus with at least one group of four reactors arranged in parallel and a dual-channel GC. An apparatus with eight reactors arranged in parallel and one or two multi-channel GCs is particularly preferred.Preferably, the multi-channel GCs are dual-channel GCs that simultaneously characterize the product fluid streams from two reactors.

[0036] A further preferred aspect of the method according to the invention is the statistical evaluation of the data, wherein this statistical evaluation is based on an evaluation model that was stored in the process control system before the catalytic test investigations were carried out. The evaluation model and the comparison between the actual and target values ​​also relate to the plausibility check of the data as well as the evaluation with regard to the process parameter to be newly set in the form of the reactor temperature. The method according to the invention is carried out such that the process parameter decisive for the process is determined by the reactor temperature of the individual reactors. In the context of the device according to the invention, it is important that the reactors can be heated individually.Individual heating of this kind is not routinely provided for test benches and is also unnecessary and useful in most cases - however, this is different for the reforming procedures at hand.

[0037] With regard to the feed fluid supply, the process according to the invention is characterized by a WHSV of 0.1 to 50 h -1< and preferably by a WHSV of 0.5 to 5 h -1<.

[0038] Regarding the control loop, it is preferred that this be an embodiment with a fuzzy logic controller. This algorithm addresses the problem of how to handle outliers within the analysis report. This includes finding and discarding peaks. The actual control algorithm is ultimately an approximation algorithm. Examples of approximation algorithms are simplex approximation, interval nesting, and similar algorithms.

[0039] The process according to the invention relates to the catalytic naphtha reforming on a laboratory scale and forms the basis for a simulation to make accurate predictions on a small scale about the behavior of catalysts and process conditions of the large-scale process. In principle, the naphtha feedstocks or starting materials are to be subjected to a refinement process through the catalytic process. Thus, the process according to the invention is designed to control the process in such a way that predetermined target properties are achieved in the product fluid stream through the presence of certain desired chemical components. The product properties play a crucial role in the evaluation and are decisive for adjusting the control parameters, whereby a precise analytical evaluation is carried out.

[0040] The procedure involves the continuous recording of the property values ​​of the target components and a suitable adjustment of the process parameters to ensure that the target properties are achieved.

[0041] The target property of the product fluid stream is to achieve a predetermined octane rating in the reformed product. Preferably, the method is designed such that the online analytical characterization is performed using a model that provides high accuracy in the evaluation.

[0042] The method according to the invention is characterized by high accuracy. Compared to the prior art, the accuracy that can be achieved with the method according to the invention is comparable to the accuracy of pilot plants that are operated as single-line plants (i.e., not in parallel) and in conjunction with the condensation of product fluids.

[0043] A method for operating a large-scale plant is described by way of example in DE 2224637. In the method of DE 2224637, the reaction effluent is analyzed. This analysis is used to determine a performance indicator and an overall conversion indicator. The performance indicator is optimized by means of an optimization device, and the reaction intensity is adjusted by means of a signal from the optimization device and a control device. The exemplary embodiment is shown for a large-scale plant that was fed with 100 m³ / h of feed material and contained 100 m³ of precious metal reforming catalyst. The products are transferred to a separator operating at low temperatures.

[0044] However, conducting investigations in (such) pilot plants involves considerably higher technical effort than the inventive method using online characterization without condensation. This accuracy is of crucial importance with regard to the usefulness of the data and the method, enabling the most precise possible statement about the catalytic properties of catalysts and processes. It is particularly important to consider that naphtha reforming is a technically demanding process, as it involves the conversion of multi-component mixtures and a multitude of components are present in the product fluid. Optimizing the large-scale process is very costly, as often only minor improvements can be achieved. However, these minor improvements are of great benefit.

[0045] An advantage of the method according to the invention is that the analytical characterization of the product fluid stream can be performed directly using the gaseous samples. It is therefore not necessary to condense the liquid product components before analysis. This significantly simplifies known methods for testing naphtha reforming. The elimination of certain apparatus components, such as the compressor and condenser, reduces the technical complexity of the apparatus. Thus, the method according to the invention is preferably characterized by online analytical characterization of the product fluid, in which prior condensation of the product fluid does not occur in at least one of the parallel reactors.

[0046] Due to the simplified apparatus in conjunction with the method according to the invention, the method is particularly suitable for use in high-throughput systems equipped with a plurality of parallel reactors. This results in a synergy effect when combining the technical design of the method according to the invention with high-throughput test facilities equipped with two or more parallel reactors.

[0047] In particular, in conjunction with the method according to the invention, it is also preferred that it be used in high-throughput plants equipped with four to 40 reactors arranged in parallel. Furthermore, it is preferred that the method according to the invention be used in conjunction with high-throughput plants equipped with four to 20 reactors arranged in parallel.

[0048] Such high-throughput systems with parallel reactors are characterized by their suitability for operation in research laboratories, with each parallel reactor being characterized by a specific catalyst capacity. The catalyst capacity of the reactors is in the range of 0.5–200 cc, preferably in the range of 1–100 cc, and more preferably in the range of 2–50 cc.

[0049] The process according to the invention is characterized by high long-term stability and can therefore be operated, for example, for a period of 10 to 100 hours; preferably, the process is operated for a period of 1 day to 180 days; more preferably, the process is operated for a period of 10 to 90 days. The particular suitability of the process according to the invention for long-term investigations is significant because it makes it possible to test the catalysts and the reforming process under conditions similar to the large-scale industrial process.

[0050] In the prior art, it is known to use comparatively expensive and large-scale pilot plants for investigating naphtha reforming. These pilot plants are operated as single-line systems with large quantities of catalysts, resulting in very high operating costs. The liquid reaction products are condensed in these pilot plants, and the knock resistance (octane number) is then determined using engine test benches.

[0051] The potential of the present method lies in the ability to test catalysts in the field of naphtha reforming in parallel under constant octane number and thus under near-industrial conditions. This can offer significant added value in the development of new catalyst systems, as well as for performance comparisons of existing catalysts.

[0052] For the method according to the invention, it is important that an online analytical characterization is carried out, wherein this online analytical characterization is performed on a gaseous product fluid stream. In particular, it is also preferred that the online analytical characterization of the gaseous product fluid stream is carried out by means of gas chromatography. Naphtha - Composition of reactant stream and product stream

[0053] The present process relates to the reaction and analysis of multicomponent mixtures containing various hydrocarbon-containing components. The term "naphtha-containing reactant fluid stream" or "naphtha-containing foreshot" refers to a mixture of hydrocarbon-containing components with 5 to 14 carbon atoms. Furthermore, the naphtha-containing reactant fluid stream also includes hydrogen, the hydrogen content of which is in the range of 0.5–60 vol.%, preferably 10–25 vol.%. The reaction of the naphtha-containing reactant fluid stream yields a reformate gasoline, which represents a naphtha-containing product fluid stream. The naphtha-containing product fluid stream can contain a variety of different hydrocarbon-containing components.For example, the naphtha-containing product fluid stream comprises 1 to 500 hydrocarbon components, preferably the naphtha-containing product fluid stream comprises 20 to 400 hydrocarbon components, more preferably the naphtha-containing product fluid stream comprises 20 to 300 hydrocarbon components.

[0054] In a preferred embodiment, the online analytical characterization is characterized by very high accuracy with regard to the qualitative and quantitative characterization of a large number of components. It should also be noted that a large number of different product components are formed, particularly during the reforming of naphtha.

[0055] The reaction of the naphtha-containing reactant fluid stream can be carried out at a pressure in the range of 1.2 - 100 bar, preferably at a pressure in the range of 1.5 - 100 bar, more preferably the process according to the invention is carried out at a pressure in the range of 2 - 35 bar.

[0056] In a preferred embodiment of the method according to the invention, the method relates to a control of the catalyst testing apparatus for the testing of catalysts during the reforming of naphtha, which is operated in such a way that the process parameters in connection with the individual reactors are adjusted so that in the product fluid streams from the individual parallel arranged reactors each exits a product fluid stream with a constant octane number (namely the C5+ fraction, i.e. the hydrocarbons with 5 or more carbon atoms).

[0057] Any technical device that outputs data correlating with an octane rating can be used for analysis. Isooctane serves as a reference, with an octane rating of 100, while heptane has an octane rating of 0. The octane rating is used to characterize the "knocking" behavior of a substance mixture.

[0058] It should be noted that the knocking behavior and the associated octane rating for individual components and groups are known in the literature, and this known data is incorporated into the inventive method. Based on the analytical characterization, the resulting calculation, and the comparison with stored data, an octane rating can be determined. The octane rating is preferably obtained by weighted summation of the quantities of individual components or component groups and their knocking properties.

[0059] Brief description of the characters Figure 1 shows an exemplary schematic of the method and control loop according to the invention, wherein an online analytical characterization of the product fluid stream from an output line is first performed, and at least one result value, preferably a plurality, is compared with at least one setpoint value, preferably a plurality. Data analysis is performed using an evaluation model, whereby process parameters are adjusted. The control loop is then repeated. Preferably, the octane number in the product fluid stream is kept constant for the duration of the catalytic test by adjusting the temperature of the individual reactors. Figure 2 shows a schematic representation of an apparatus with four reactors arranged in parallel, which are equipped with an online analysis unit, in which the analysis unit is operatively connected to a partial flow line from output line 1.Figure 3 shows a schematic representation of the octane number values ​​(maintained constant while the temperatures of the different catalysts are installed in three parallel reactors) on the upper part of the y-axis. Measurements were taken over a period of 10 hours. The lower part of the y-axis shows the temperature tracking of the respective reactor heaters for each catalyst. The x-axis reflects the time period required to maintain the octane number of the C5+ hydrocarbons at a constant level. Figure 4 shows a schematic of a controller coupled to a system. "System" here refers to the parallel reactor apparatus. The controller receives time-dependent values ​​from the product fluid flows in the individual lines and then adjusts a process parameter. Figure 5 shows an overview of a control loop that is part of a catalyst testing apparatus.According to the method according to the invention, the control loop is part of an apparatus with reactors arranged in parallel. Figure 6 shows an overview of a control loop in a process sequence, where the division between process 1 and process 2 is shown, each representing two different reactor units of a catalyst test apparatus.

[0060] In all illustrated embodiments of the Figures 4 to 6 The coupling of the process control refers to the connection with a parallel reactor apparatus.

[0061] The term "system" has no restrictive effect here and applies to any arrangement of parallel reactors. The term "process" also refers to... Figure 5 on the process in combination with parallel reactors.

[0062] The number of reactors arranged in parallel is limited by technical feasibility with regard to upper limits concerning the number of reactors. One aspect of the method according to the invention is the use of two or more parallel reactors, preferably four or more. The degree of parallelization influences the design of the method, since the analysis time for product fluids is subject to certain minimum times. In addition to the analysis time, other parameters also influence the runtime. In particular, it is preferred that the analysis column is regenerated after the analysis has been carried out.

[0063] This results in processing times (runtimes) for the groups process analytics, analysis evaluation, and data processing. The runtime in these groups is preferably no more than half the system's response time (Nyquist-Shannon sampling theorem).

[0064] In this context, the term "multiplexer" refers to a selection valve that sequentially switches the product fluid flows from the respective reactors to the analyzer (which is also referred to as "multiplexed"). "Demultiplexing" the controller means that the manipulated variable determined from the data is assigned to the reactor—and the process occurring within that reactor—that originally provided that data. It follows that the multiplexer and demultiplexer must have the same number of positions.

[0065] The following describes individual process steps of the process according to the invention as described in the Figure 1 , 5 and 6 shown and explained in more detail: 1. A multi-component mixture from a reactor outlet line is separated into its individual components using an online analysis unit. 2. The separated individual components exhibit more or less pronounced differences in retention times, which are associated with temporal trends. a. Preferably, the analytical evaluation is linked to a grouping of the analytical data according to boiling ranges, and the result of this evaluation is preferably used to determine the octane rating. Data compression to a single value is particularly preferred. It is necessary that the actual controller is supplied with only a time-dependent input variable. Furthermore, it is preferred that the data compression is model-based.In connection with the method according to the invention, the model consists of condensing the analytical characterization data, preferably determined by online gas chromatography, into an octane number. It is also conceivable that the analytical characterization is carried out using IR spectroscopy. 3. The controller is an element of the apparatus that reacts to a parameter over time. a. In this specific case, the controller is in particular: i. implemented programmatically (is an algorithm) ii. designed as an interval-controlled search and approximation algorithm. iii. Alternative: simplex search algorithm; other search algorithms are also covered by the present invention.

[0066] Process analytics, in the case of chromatographic methods, creates discrete time-based representations of the process. At specific time intervals, the sampling frequency, individual samples of the substance mixtures (i.e., the reformates) exiting the respective reactor are taken.

[0067] The substance mixtures are analyzed, i.e., separated into their components. In naphtha reforming and the process according to the invention, it is particularly preferred to perform a gas chromatographic analysis. Gas chromatographic analysis can be considered an incomplete chromatography, since the substance mixture is separated into groups whose mutual delimitation is achieved by characteristic, known substances.

[0068] Thus, it is an integral part of the inventive method that several time trends are compressed into a single time trend. Preferably, the actual value of the octane rating is entered into the controller (specifically as a function x = octane rating(t)). Instead of the term octane rating, the term Research Octane Number, abbreviated as RON, can also be used.

[0069] A controller according to the present invention can process exactly one input variable. Therefore, if the system to be controlled provides multiple actual values, these must be condensed into a single time-dependent value.

[0070] A controller can control exactly one system. The controller's response (manipulated variable) can be distributed across multiple actuators, thus enabling multiple control options. For example, the temperature and flow rate of a reactor can be connected to the same controller output, and therefore the same manipulated variable. The system then responds to all influencing variables simultaneously.

[0071] Preferably, two or more controllers are used to influence the temperature of a reactor. There are preferably as many systems as there are controllers. (Thermosensor(1); Controller(2); Actuator(3); Heating element(4); Thermosensor(1)). For the control system(s) to function correctly, it must be ensured that the mutual interference between the controllers is sufficiently low. If a certain threshold for mutual interference is exceeded, oscillations of the systems can occur. These oscillations are generally undesirable, as they can lead to critical operating conditions. Examples

[0072] To illustrate the process according to the invention, a test run was carried out with five different catalysts under naphtha reforming conditions. Industrially produced catalysts were used, approximately 9 g of catalyst per reactor. The duration of the investigation was 200 hours, with the results presented in the Figure 3 The graph shows a section of a 10-hour period. The individual experiments were conducted in such a way that the octane rating was kept constant, which is evident in the upper part of the graph. Figure 3This can be seen. The temperature of the individual reactors was continuously increased to compensate for the loss of activity observed in connection with catalyst aging. On average, the temperature was increased by 1.5 to 5°C every 200 hours to maintain the octane number of the product fluid stream within a constant range. The measured octane number values ​​show a standard deviation of ±0.3. It should be noted that this is a very low standard deviation, which has not been reported in the literature for laboratory equipment. This high accuracy is of great significance with regard to the technical relevance of the data obtained using the process according to the invention. Reference symbol list

[0073] 1- Reactor 1 1iv- Reactor 4 2- Output line in connection with Reactor 1 2iv Output line in connection with Reactor 4 3 Partial power line 4- Online analysis unit 5 Process control 6- Reactor heating to Reactor 1 6iv- Reactor heating to Reactor 4 7- Connection of process control with reactor heating 1

Claims

1. An apparatus for investigating naphtha reforming, the apparatus comprising: a plurality of reactors arranged in parallel with reaction chambers (R1, R2, ···), the individual reactors having a catalyst capacity in the range from 0.5 to 200 ccm, a reactant fluid supply common to two or more of these reactors, a process control, an analysis unit, where each individual reactor is equipped with an individually controllable heating device and has an outlet line for product fluid stream, wherein the analysis unit is operatively connected to each said outlet line for product stream, and wherein the analysis unit is functionally associated with the control of the apparatus, where the analysis unit is suitable for analyzing a gaseous product fluid stream, and the analysis unit together with the process control allow the determination and optimization of an octane number in the reactors arranged in parallel, where the analysis unit is a gas chromatograph or the gaseous product fluid stream is characterized by means of IR spectroscopy, and where the temperature of the controllable heating device of the respective reactor from whose outlet line the product fluid stream was analyzed can be adapted.

2. The apparatus according to claim 1, wherein the reactor is a tubular reactor.

3. The apparatus according to either of the preceding claims, wherein the apparatus has a number of 2 to 40 reactors arranged in parallel.

4. The apparatus according to any of the preceding claims, wherein a temperature sensor is arranged in the interior of at least one of the reactors arranged in parallel.

5. The apparatus according to any of the preceding claims, wherein the outlet lines of the individual reactors are connected to a pressurizing gas unit and a pressure-maintaining gas unit.

6. The apparatus according to any of the preceding claims, wherein the outlet lines of the individual reactors are equipped with membrane valves and the pressure within the reactors is controlled by the setting at the membrane valve.

7. A method for investigating naphtha reforming using an apparatus according to claims 1 to 6, where a naphtha-containing reactant fluid stream comprising: a mixture of hydrocarbons having 5 to 14 carbons, and hydrogen with a hydrogen content in the range of 0.5-60% by volume, is supplied to the individual reactors R1, R2,..., and is contacted in the respective reactors with a catalyst at a temperature in the range from 100°C to 600°C, and converted to one product fluid stream each, and this product fluid stream is subsequently supplied to an outlet line, wherein the method is characterized by the following method steps: (i) an analytical characterization of the individual product fluid streams by means of an analysis unit, the analysis relating to the quantification of one or more components, (ii) comparison of the result value of the quantification with a set value, where the comparison comprises the storage of an evaluation model, where the evaluation model relates to the calculation of the octane number, (iii) adaptation of the temperature of the controllable heating device of the respective reactor from whose outlet line the product fluid stream was analyzed, (iv) repetition of method steps (i) through (iii) by analysis of the product fluid stream emerging from the respective outlet lines.

8. The method for investigating naphtha reforming according to claim 7, wherein the duration for the analytical characterization is in a time range from 0.1 min to 300 min.

9. The method for investigating naphtha reforming according to either of claims 7 and 8, wherein the method is carried out over a duration of 24 h to 3000 h.

10. The method for investigating naphtha reforming according to any of claims 7 to 9, wherein the method is conducted such that the octane number is adapted such that the octane number of the reformate in the product fluid stream has a predetermined target value.

11. The method for investigating naphtha reforming according to any of claims 7-10, wherein the contacting of the naphtha-containing reactant fluid stream is carried out at a pressure in the range from 1.2 to 100 bar.