Method and system for simulating and correcting reaction for preparing styrene through ethylbenzene dehydrogenation

Abstract

The invention relates to a method for simulating and correcting reaction for preparing styrene through ethylbenzene dehydrogenation. The method comprises the following steps: step 1) establishing a reaction kinetic equation of a chemical reaction; step 2) establishing a material balance equation of each reactant and reaction product according to the reaction kinetic equation in the step 1); 3) establishing an energy balance equation according to the material balance equation in the step 2); 4) establishing a momentum balance equation according to the energy balance equation; 5) establishing an ethylbenzene dehydrogenation radial adiabatic reactor model by adopting the reaction kinetic equation, the material balance equation, the energy balance equation and the momentum balance equation; 6) establishing a correction target function, then establishing a batch gradient descent algorithm required by correction calculation, and correcting constants in the ethylbenzene dehydrogenation radial adiabatic reactor model by taking minimization of the correction target function as a target; and performing simulation calculation through the corrected ethylbenzene dehydrogenation radial adiabatic reactor model to obtain a corrected simulation result.

Description

Technical Field

[0001] The invention relates to the technical field of chemical engineering, and in particular to a simulation and correction method and system for a reaction of ethylbenzene dehydrogenation to produce styrene. Background Art

[0002] Styrene is one of the most important monomers in modern petrochemical products. It is mainly used to produce polystyrene, unsaturated polyester, ABS / SAN resin, styrene-butadiene rubber and latex, styrene series thermoplastic elastomers, etc. Ethylbenzene dehydrogenation to styrene is the main method for producing styrene. In industrial applications, the reaction process is usually carried out under negative pressure and water vapor. In addition to the main reaction of ethylbenzene dehydrogenation to styrene, it is also accompanied by various side reactions such as thermal cracking and hydrogenation cracking. These side reactions will produce toluene, benzene, methane and ethylene. Among them, methane and ethylene will further undergo water vapor conversion reaction to produce gases such as carbon monoxide and carbon dioxide. In the ethylbenzene dehydrogenation process, some impurities in the raw material ethylbenzene will also participate in the reaction. For example, isopropylbenzene can produce α-methylstyrene during the dehydrogenation process. In view of the complexity of the ethylbenzene dehydrogenation reaction network, it is particularly important to use mathematical methods to automatically calculate and correct the reaction kinetic constants based on experimental data and production data. Many scholars have conducted in-depth research in this field.

[0003] However, the current work on the correction of the kinetic constants of the dehydrogenation of ethylbenzene to styrene involves a relatively simple reaction network and only stays at the level of kinetic parameter correction. The actual dehydrogenation reaction of ethylbenzene is relatively complex, and industrial reactors cannot be strictly adiabatic. The pressure calculation is also inconsistent with the pressure drop calculation formula such as the Ergun equation due to the difference between the catalyst particles and the stacking state, so the scope of application is limited and it is not universally applicable. Summary of the invention

[0004] The present invention proposes a correction method and system for the reaction model of ethylbenzene dehydrogenation to styrene in the ST process, including the rate constants of one main reaction and eight side reactions of ethylbenzene dehydrogenation to styrene and the temperature drop and pressure drop of two adiabatic reactors in the ST process, totaling 13 correction constants, and the batch gradient descent method is used to solve the objective function. This method considers more detailed correction items for the industrial ethylbenzene dehydrogenation reaction process, ensures the accuracy of the simulation prediction of the actual ethylbenzene dehydrogenation to styrene reaction system, and can optimize the operation of the reaction system and improve the efficiency of the device. The present invention is implemented by the following technical solutions:

[0005] A simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene comprises the following steps:

[0006] Step 1) Establish the reaction kinetics equation of the following chemical reaction:

[0007] Reaction 1: Ethylbenzene reacts to produce styrene and hydrogen;

[0008] Reaction 2: Ethylbenzene reacts to produce benzene and ethylene;

[0009] Reaction 3: Ethylbenzene and hydrogen react to form toluene and methane;

[0010] Reaction 4: Styrene and hydrogen react to produce benzene and ethylene;

[0011] Reaction 5: Styrene and hydrogen react to produce toluene and methane;

[0012] Reaction 6: Methane and water react to produce carbon monoxide and hydrogen;

[0013] Reaction 7: Carbon monoxide reacts with water to produce carbon dioxide and hydrogen;

[0014] Reaction 8: Ethylene and water react to produce carbon monoxide and hydrogen;

[0015] Reaction 9: Cumene reacts to produce α-methylstyrene and hydrogen;

[0016] Step 2) establishing a material balance equation for each reactant and reaction product according to the reaction kinetic equation in step 1);

[0017] Step 3) establishing an energy balance equation based on the material balance equation in step 2);

[0018] Step 4) establishing a momentum balance equation based on the material balance equation in step 2 and the energy balance equation in step 3);

[0019] Step five) using the reaction kinetics equation, material balance equation, energy balance equation and momentum balance equation, according to the content of each component in the raw material fed into the reactor, the raw material temperature, the raw material pressure, and the reactor structural parameters, the simulation calculation obtains the content of each component in the reaction product output by the reactor, the reaction product temperature, and the reaction product pressure.

[0020] Step six) establishing a correction objective function, and then establishing a batch gradient descent algorithm required for the correction calculation, with the goal of minimizing the correction objective function, correcting the model constant of the radial adiabatic reactor for ethylbenzene dehydrogenation; simulating the corrected ethylbenzene dehydrogenation radial adiabatic reactor model to obtain the corrected content of each component in the reaction product, the reaction product temperature, and the reaction product pressure; the correction objective function is:

[0021]

[0022] Where n is the number of system components, m is the number of reactors, x cal is the simulated value of the component molar fraction, x real is the target value of component molar fraction, w iis the component weight factor, w j is the reactor temperature drop / pressure drop weight factor, ΔT cal is the simulated value of reactor temperature drop, ΔT real is the target value of reactor temperature drop, ΔP cal is the simulated value of reactor pressure drop, ΔP real is the target value of reactor pressure drop.

[0023] The radial adiabatic reactor model constants for ethylbenzene dehydrogenation include the reaction rate constants of reactions 1 to 9 and the temperature drop and pressure drop of the adiabatic reactors (two reactors in the present invention), totaling 13 calibration constants.

[0024] Optionally, the reaction kinetics equation includes:

[0025]

[0026] K P =C1×exp(C2-C3 / T-C4×log(T))

[0027] Where r is the reaction rate, unit is kmol / (kg·h); Kp is the main reaction equilibrium constant, unit is kPa; C1~C4 are correlation constants; k i is the reaction rate constant, unit kmol / (kg·h·kPa m ·K n ), where m is determined by dimensional analysis and is equal to 1-total number of series, n is only used by R7 and has a value of 3; k i0 is the reaction rate pre-exponential factor, with the same unit as k i Same; E a is the activation energy, in J / mol; b j is the adsorption constant, unit kPa -1 ; b j0 is the adsorption pre-exponential factor, unit kPa -1 ;Q j is the adsorption activation energy, in J / mol. ST is the partial pressure of styrene, P EB is the partial pressure of ethylbenzene, P H2O is the water partial pressure, P H2 is the hydrogen partial pressure, P C2H4 is the ethylene partial pressure, P CH4 is the methane partial pressure, P CO is the partial pressure of carbon monoxide, P ISB is the partial pressure of cumene.

[0028] Optionally, the material balance equation is:

[0029]

[0030] The flow rate change rate of impurity components that do not participate in the reaction, such as p-xylene, p-diethylbenzene and non-aromatics, is 0:

[0031]

[0032] Where F is the molar flow rate of the component, in kmol·h -1 , r i is the reaction rate, in kmol kg -1 ·h -1 , i = 1 to 9, r is the bed radius, unit: m, L is the bed height, unit: m, ρ is the catalyst bulk density, unit: kg·m -3 , Other represents impurities. ST is the molar flow rate of styrene, F TL is the molar flow rate of toluene, F EB is the molar flow rate of ethylbenzene, F BZ is the molar flow rate of benzene, F H2O is the water molar flow rate, F H2 is the hydrogen molar flow rate, F C2H4 is the molar flow rate of ethylene, F CH4 is the molar flow rate of methane, F CO is the molar flow rate of carbon monoxide, F CO2 is the molar flow rate of carbon dioxide, F ISB is the molar flow rate of cumene, F AMS is the molar flow rate of α-methylstyrene, F other It is an impurity that does not participate in the reaction.

[0033] Optionally, the energy balance equation is:

[0034]

[0035] Where ΔH j is the enthalpy change of each reaction, j = 1 to 9, unit J·mol -1 , C pi is the molar heat capacity at constant pressure, i = 1 to 19, unit J·mol -1 ·K -1 .

[0036] Optionally, the momentum balance equation is:

[0037]

[0038] Where ε is the bed porosity, d p is the equivalent diameter of the catalyst particle specific surface area, in m, μ is the dynamic viscosity of the mixed gas, in Pa·s, and u is the gas superficial velocity, in m·s -1 .

[0039] Optionally, the reactor in the step five) includes a first reactor and a second reactor connected in series, and the reaction kinetics equation, material balance equation, energy balance equation and momentum balance equation are used to sequentially calculate the reaction product temperature, reaction product pressure and reaction product composition of the first reactor and the second reactor.

[0040] Optionally, a heat exchanger is provided on the second reactor, and the heat exchange temperature and heat exchange load of the heat exchanger are calculated using the following formula:

[0041]

[0042] Where T in,steam is the steam inlet temperature, in K, T out,steam is the steam outlet temperature, unit K, F steam is the steam molar flow rate, unit is kmol·h -1 , C p,steam is the molar heat capacity of steam at constant pressure, in J·mol -1 ·K -1 , T in,Material is the reaction material inlet temperature, unit K, T out,Material is the outlet temperature of the reaction material, unit K, F Material is the molar flow rate of the reaction material, unit is kmol·h -1 , C p,Material is the molar heat capacity of the reaction materials at constant pressure, in J·mol -1 ·K -1 .

[0043] Optional,

[0044] The batch gradient descent algorithm required for establishing the correction calculation is used to correct the model constants of the radial adiabatic reactor for ethylbenzene dehydrogenation with the goal of minimizing the correction objective function. The calculation steps are as follows:

[0045] Step 1: Input the parameters required for calibration to form the initial task;

[0046] Step 2: Add the initial task to the task queue and input the batch gradient descent method convergence parameters: minimum learning rate, minimum branch rate, error threshold eps, number of single-round iterations k, and initial learning rate α;

[0047] Step 3: If the task queue is not empty, select the task with the smallest error f in the queue for optimization. If the queue is empty, the iteration fails and the program ends.

[0048] Step 4: Given the initial point x (0) ∈R n , calculate the error function f;

[0049] Step 5: Calculate the (k ) The gradient g at k = Δf(x (k) );

[0050] Step 6. Let d (k) = -g k ;

[0051] Step 7. Starting from x (k) , perform a one-dimensional search along the direction d (k) , and let x (k+1) = x (k) + α·d (k) ;

[0052] Step 8. Check whether the convergence criterion is met. If f < eps, stop the iteration, and the approximate minimum point is x (k) , otherwise go to step 9;

[0053] Step 9. If f is less than the branch threshold, create a new task with the same number of iterations and a learning rate halved, and add it to the task queue;

[0054] Step 10. Let k = k + 1. If k is equal to the written number of single-round iterations, stop the iteration, indicating that this round of iteration fails, and return to step 3; otherwise return to step 5;

[0055] Preferably, when running the batch gradient descent algorithm, input the solution parameters and calibration calculation parameters of the two-stage ethylbenzene dehydrogenation radial adiabatic reactor model with an intermediate heat exchanger;

[0056] The solution parameters of the two-stage ethylbenzene dehydrogenation radial adiabatic reactor model with an intermediate heat exchanger include: the initial values of the kinetic parameters of each reaction; the reactor structure parameters and catalyst specifications of the styrene plant, including the inner diameter, outer diameter, height, porosity, catalyst bulk density, and catalyst equivalent diameter of the catalyst bed in the first reactor, and the inner diameter, outer diameter, height, porosity, catalyst bulk density, and catalyst equivalent diameter of the catalyst bed in the second reactor; the industrial feed flow rate, temperature, pressure, and composition; the temperature, pressure, and flow rate of the steam in the intermediate heat exchanger; the temperature and pressure after heat exchange of the raw material through the intermediate heat exchanger;

[0057] The calibration calculation parameters include: the flow rate, temperature, pressure, and composition of the dehydrogenated product in actual production in the same period as the feed data; the inlet and outlet temperatures and pressures of the reactor; and the calibration weight factors for the temperature drop and pressure drop of each component and each reactor.

[0058] A simulation and calibration system for the ethylbenzene dehydrogenation to styrene reaction, comprising:

[0059] A reaction kinetics module for calculating the reaction rates of each reaction in the reactor;

[0060] Reaction material balance module, used to calculate the flow and composition of each reaction raw material and reaction product in the reactor;

[0061] Reaction energy balance module, used to calculate the temperature change and reaction heat in the reactor;

[0062] Reaction momentum balance module, used to calculate the pressure change in the reactor;

[0063] A correction calculation module is used to correct the reaction model constants and simulate the content of each component in the reaction product, the temperature of the reaction product, and the pressure of the reaction product after correction;

[0064] The reaction of the reactor includes:

[0065] Ethylbenzene reacts to produce styrene and hydrogen;

[0066] Ethylbenzene reacts to produce benzene and ethylene;

[0067] Ethylbenzene reacts with hydrogen to produce toluene and methane;

[0068] Styrene reacts with hydrogen to produce benzene and ethylene;

[0069] Styrene reacts with hydrogen to produce toluene and methane;

[0070] Methane and water react to form carbon monoxide and hydrogen;

[0071] Carbon monoxide reacts with water to form carbon dioxide and hydrogen;

[0072] Ethylene and water react to form carbon monoxide and hydrogen;

[0073] Cumene reacts to produce α-methylstyrene and hydrogen.

[0074] Optionally, the reactor includes a first reactor and a second reactor connected in series.

[0075] Optionally, a heat exchanger is provided on the second reactor, and the system further comprises:

[0076] The heat exchange calculation module is used to calculate the heat exchange temperature and heat exchange load of the heat exchanger.

[0077] Compared with the prior art, the present invention has the following advantages and effects:

[0078] The present invention proposes a simulation and correction method for the ST process ethylbenzene dehydrogenation to styrene reaction. According to the reaction mechanism of ethylbenzene dehydrogenation to styrene and the structural type of the ST process ethylbenzene dehydrogenation to styrene reactor, a reaction kinetics model and a reactor model including 1 main reaction and 8 side reactions are constructed, and temperature drop and pressure drop correction factors for each reaction and each reactor are introduced. Based on the reaction kinetics model, the reactor model and the batch gradient descent algorithm, the reaction kinetics model and the reactor model can be corrected to ensure the accuracy of the simulation prediction of the ethylbenzene dehydrogenation reaction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 It is a structural diagram of the reactor of the present invention. DETAILED DESCRIPTION

[0080] In order to make the purpose and technical solution of the present invention more clear, the present invention is further described in detail. The experimental methods described in the following examples are all conventional methods unless otherwise specified: if no specific techniques or conditions are specified in the examples, they are carried out according to the techniques or conditions described in the literature in the field or according to the product instructions; the reagents and materials, unless otherwise specified, can be obtained from commercial channels.

[0081] Example 1

[0082] The present invention provides a simulation and correction method for the reaction of ethylbenzene dehydrogenation to produce styrene, comprising the following steps:

[0083] Step 1) Establish the reaction kinetics equation of the following chemical reaction:

[0084] Reaction 1: Ethylbenzene reacts to produce styrene and hydrogen;

[0085] Reaction 2: Ethylbenzene reacts to produce benzene and ethylene;

[0086] Reaction 3: Ethylbenzene and hydrogen react to form toluene and methane;

[0087] Reaction 4: Styrene and hydrogen react to produce benzene and ethylene;

[0088] Reaction 5: Styrene and hydrogen react to produce toluene and methane;

[0089] Reaction 6: Methane and water react to produce carbon monoxide and hydrogen;

[0090] Reaction 7: Carbon monoxide reacts with water to produce carbon dioxide and hydrogen;

[0091] Reaction 8: Ethylene and water react to produce carbon monoxide and hydrogen;

[0092] Reaction 9: Cumene reacts to produce α-methylstyrene and hydrogen;

[0093] Step 2) establishing a material balance equation for each reactant and reaction product according to the reaction kinetic equation in step 1);

[0094] Step 3) establishing an energy balance equation based on the material balance equation in step 2);

[0095] Step 4) establishing a momentum balance equation based on the material balance equation in step 2 and the energy balance equation in step 3);

[0096] Step five) using the reaction kinetics equation, material balance equation, energy balance equation and momentum balance equation, according to the content of each component in the raw material fed into the reactor, the raw material temperature, the raw material pressure, and the reactor structural parameters, the simulation calculation obtains the content of each component in the reaction product output by the reactor, the reaction product temperature, and the reaction product pressure.

[0097] Step six) establishing a correction objective function, and then establishing a batch gradient descent algorithm required for the correction calculation, with the goal of minimizing the correction objective function, correcting the model constants of the radial adiabatic reactor for ethylbenzene dehydrogenation; using the corrected radial adiabatic reactor model for ethylbenzene dehydrogenation, simulating a more accurate content of each component in the reaction product, the reaction product temperature, and the reaction product pressure; the correction objective function is:

[0098]

[0099] Where n is the number of system components, m is the number of reactors, x cal is the simulated value of the component molar fraction, x real is the target value of component molar fraction, w i is the component weight factor, w j is the reactor temperature drop / pressure drop weight factor, ΔT cal is the simulated value of reactor temperature drop, ΔT real is the target value of reactor temperature drop, ΔP cal is the simulated value of reactor pressure drop, ΔP real is the target value of reactor pressure drop.

[0100] The radial adiabatic reactor model constants for ethylbenzene dehydrogenation include the reaction rate constants of reactions 1 to 9 and the temperature drop and pressure drop of the adiabatic reactors (two reactors in the present invention), totaling 13 calibration constants.

[0101] Reactor Figure 1 The figure shows a two-stage series radial reactor model, which includes a first reactor and a second reactor in series, wherein a heat exchanger is also provided at the inlet of the second reactor.

[0102] Reactions considered include:

[0103]

[0104] EB→BZ+C2H4 (2)

[0105] EB+H2→TL+CH4 (3)

[0106] ST+H2→BZ+C2H4 (4)

[0107] ST+2H2→TL+CH4 (5)

[0108] CH4+H2O→CO+3H2 (6)

[0109] CO+H2O→CO2+H2 (7)

[0110] 0.5C2H4+H2O→CO+2H2 (8)

[0111] ISB→AMS+H2 (9)

[0112] Among them, EB is ethylbenzene, ST is styrene, TL is toluene, BZ is benzene, ISB is isopropylbenzene, and AMS is α-methylstyrene.

[0113] The reaction kinetics equation includes:

[0114]

[0115]

[0116] K P =C1×exp(C2-C3 / T-C4×log(T)) (21)

[0117] Where r is the reaction rate (kmol / (kg·h)), Kp is the main reaction equilibrium constant (kPa), C1~C4 are correlation constants, C1~C4 are correlation constants, which are obtained from the existing public ethylbenzene dehydrogenation to styrene reaction data according to the reaction conditions such as the catalyst used, and k i is the reaction rate constant (kmol / (kg·h·kPa m ·K n ) where m is determined by dimensional analysis and is equal to 1-total number of series, n is only used by r7 and has a value of 3), k i0 is the reaction rate pre-exponential factor, with the same unit as k i Same, E a is the activation energy (J / mol), b j is the adsorption constant (kPa -1 ), b j0 is the adsorption pre-exponential factor, Q j is the adsorption activation energy (J / mol).

[0118] The ethylbenzene dehydrogenation reactor is a two-stage series radial reactor with intermediate heat exchange. The reactor structure is as follows Figure 1 As shown, the catalyst is loaded in the annular area of ​​the reactor. After the ethylbenzene vapor is heated by heat exchange, it is mixed with the superheated water vapor at the inlet of the first dehydrogenation reactor and enters the catalyst bed along the axial radial direction. The dehydrogenation reaction occurs under negative pressure adiabatic conditions. Since the dehydrogenation of ethylbenzene is an endothermic reaction, the outlet temperature of the first dehydrogenation reactor decreases. After the temperature is increased by heat exchange with the superheated water vapor in the intermediate heat exchanger, it enters the catalyst bed of the second dehydrogenation reactor to realize the two-stage negative pressure adiabatic dehydrogenation reaction. The logistics discharged from the outlet of the second dehydrogenation reactor is the final reaction product.

[0119] The mathematical model equation group of the ethylbenzene dehydrogenation reactor includes a material balance equation, a heat balance equation, a momentum balance equation and a heat balance equation for an intermediate heat exchanger.

[0120] The material balance equation is:

[0121]

[0122]

[0123] The flow rate change rate of impurity components that do not participate in the reaction, such as p-xylene, p-diethylbenzene and non-aromatics, is 0:

[0124]

[0125] Where F is the molar flow rate (kmol / h), r i is the reaction rate (kmol kg -1 ·h -1 ), i = 1 to 9, r is the bed radius (m), L is the bed height (m), ρ is the catalyst bulk density (kg / m 3 ), Other represents impurities, specifically including: o-xylene, m-xylene, p-xylene, o-diethylbenzene, m-diethylbenzene, p-diethylbenzene and non-aromatics.

[0126] The energy balance equation is:

[0127]

[0128] Where ΔH j is the enthalpy change of each reaction, j = 1 to 9, unit J·mol -1 , C pi is the molar heat capacity at constant pressure, i = 1 to 19, unit J·mol -1 ·K -1, the energy balance considers 9 reactions and 19 components (including reactants and major impurities, where i from 1 to 19 represents ethylbenzene, styrene, toluene, benzene, water, hydrogen, ethylene, methane, carbon monoxide, carbon dioxide, cumene, α-methylstyrene, o-xylene, m-xylene, p-xylene, o-diethylbenzene, m-diethylbenzene, p-diethylbenzene, and non-aromatics). The heat of each reaction is as follows:

[0129] ΔH1=-120697-4.56T / K

[0130] ΔH2=-108750+7.95T / K

[0131] ΔH3=53145+13.18T / K

[0132] ΔH4=11929+12.51T / K

[0133] ΔH5=173824+17.74T / K

[0134] ΔH6=-206100

[0135] ΔH7=41000

[0136] ΔH8=-225600

[0137] ΔH9=-119554

[0138] The constant pressure heat capacity equations of each component and mixture are as follows:

[0139] C P,i =A i +B i ×T+C i ×T 2 +D i ×T 3

[0140]

[0141] where y i is the mole fraction, and A, B, C, and D are correlation constants derived from a public database.

[0142] Momentum loss is based on the pressure drop formula for fluid flowing in an empty circular tube, specifically:

[0143]

[0144] Where ε is the bed porosity, d p is the equivalent diameter of the catalyst particle specific surface area, in m, μ is the dynamic viscosity of the mixed gas, in Pa·s, and u is the gas superficial velocity, in m·s -1, the viscosity equations of each component and mixture are as follows:

[0145]

[0146] Among them, μ i is the viscosity of the pure component, in Pa·s, C i is the viscosity correlation constant, which comes from a public database, i = 1~4, μ mix is the viscosity of the mixture, in Pa·s, M i is the relative molecular mass, i = 1 to 19, unit: g·mol -1 .

[0147] The heat balance equation of the intermediate heat exchanger is:

[0148]

[0149] Where T in,steam is the steam inlet temperature, in K, T out,steam is the steam outlet temperature, unit K, F steam is the steam molar flow rate, unit is kmol·h -1 , C p,steam is the molar heat capacity of steam at constant pressure, in J·mol -1 ·K -1 , T in,Material is the reaction material inlet temperature, unit K, T out,Material is the outlet temperature of the reaction material, unit K, F Material is the molar flow rate of the reaction material, unit is kmol·h -1 , C p,Material is the molar heat capacity of the reaction materials at constant pressure, in J·mol -1 ·K -1 .

[0150] A correction objective function is established. The correction calculation takes the minimum objective function as the optimization target. The objective function is the calculated value of the composition of each component at the reactor outlet (the molar fraction of ethylbenzene, styrene, toluene, benzene, ethylene, methane, carbon monoxide, carbon dioxide, isopropylbenzene, and α-methylstyrene), the temperature drop of the first reactor and the second reactor, and the sum of the squares of the pressure drop and the actual values ​​of these quantities, in the following form: Where n is the number of system components, m is the number of reactors, x cal is the simulated value of the component molar fraction, x real is the target value of component molar fraction, w i is the component weight factor, w j is the reactor temperature drop / pressure drop weight factor, ΔT cal is the simulated value of reactor temperature drop, ΔT realis the target value of the reactor temperature drop, ΔP cal is the simulated value of the reactor pressure drop, ΔP real is the target value of the reactor pressure drop.

[0151] Establish the batch gradient descent algorithm required for calibration calculation. With the goal of minimizing the calibration objective function, calibrate the model constants of the ethylbenzene dehydrogenation radial adiabatic reactor. The calculation steps are as follows:

[0152] 1. Input the parameters required for calibration to form an initial task.

[0153] 2. Add the initial task to the task queue, and input the minimum learning rate, minimum branching rate, error threshold eps, number of iterations per round k, and initial learning rate α.

[0154] 3. When the task queue is not empty, select the task with the smallest error f in the queue for optimization (the queue has been determined to have the smallest f first through the sorting algorithm). If the queue is empty, the iteration fails and the program ends.

[0155] 4. Given the initial point x (0) ∈R n , calculate the error function f.

[0156] 5. Calculate the gradient at x ( k ) .

[0157] 6. Let d (k) =-g k .

[0158] 7. Starting from x (k) , perform a one-dimensional search along the direction d (k) , and let x (k+1) =x (k) +α·d (k) .

[0159] 8. Check whether the convergence criterion is met. If f < eps, stop the iteration, and the approximate minimum point is x (k) , otherwise go to 9.

[0160] 9. If f is less than the branching threshold (built-in), create a new task with the same number of iterations and a learning rate halved, and add it to the task queue.

[0161] 10. Let k = k + 1. If k is already equal to the number of iterations per round written, stop the iteration, indicating that this round of iteration fails, and return 3; otherwise return 5.

[0162] When running the batch gradient descent algorithm, the solution parameters and calibration calculation parameters of the two-stage ethylbenzene dehydrogenation radial adiabatic reactor with an intermediate heat exchanger should be input simultaneously.

[0163] The solution parameters of the two-stage ethylbenzene dehydrogenation radial adiabatic reactor model with an intermediate heat exchanger include: the initial values ​​of the kinetic parameters of 9 reactions; the structural parameters of the styrene unit reactor and the catalyst specifications, including the inner diameter, outer diameter, height, porosity, catalyst stacking density, and catalyst equivalent diameter of the catalyst bed of the first reactor, and the inner diameter, outer diameter, height, porosity, catalyst stacking density, and catalyst equivalent diameter of the catalyst bed of the second reactor; the flow rate, temperature, pressure, and composition of industrial feed; the temperature, pressure, and flow rate of the steam of the intermediate heat exchanger; and the temperature and pressure of the raw material after exchange through the intermediate heat exchanger.

[0164] The correction calculation parameters include: the flow rate, temperature, pressure and composition of the industrial dehydrogenation product in the same period as the industrial feed data; the reactor inlet and outlet temperature and pressure; the correction weight factors of the temperature drop and pressure drop of each component and each reactor; batch gradient descent method convergence parameters: minimum learning rate, minimum branching rate, error threshold eps, single round iteration number k, initial learning rate α.

[0165] Test example:

[0166] The above method is used to simulate and calibrate the styrene unit. The reactor structural parameters and catalyst specifications of the styrene unit are shown in Table 1; the temperature, pressure, flow rate and composition of the industrial feed are shown in Table 2; the data of the intermediate heat exchanger and the raw materials after the intermediate heat exchanger are shown in Table 3.

[0167] Table 1 Reactor structure and catalyst specifications

[0168]

[0169]

[0170] Table 2 Industrial feed data

[0171] Logistics Name Feed ethylbenzene vapor Feed superheated steam Temperature, °C 502.7 834 Pressure, Kpa 43.2 238.3 Flow rate, kg / h 25094.6 13700 Composition, mass TOL 0.000877 EB 0.657509 STY 0.003549 CUM 0.000119 PX 0.003212 H2O 0.334734 1

[0172] Table 3 Industrial operation data

[0173] project unit Numeric Second reactor inlet temperature ℃ 615 Second reactor inlet pressure Kpa 30 Intermediate heat exchanger steam inlet temperature ℃ 814 Steam inlet pressure of intermediate heat exchanger Kpa 250 Steam mass flow rate of intermediate heat exchanger kg / h 13700

[0174] The product flow and composition data after the dehydrogenation reaction are shown in Table 4; the reactor temperature drop and pressure drop correction targets are shown in Table 5; the temperature drop and pressure drop correction weights of each component and reactor are shown in Table 6; and the convergence parameters of the batch gradient descent method are shown in Table 7.

[0175] Table 4 Reaction product flow and composition

[0176]

[0177]

[0178] Products 1 to 4 in Table 6 are input by the user, and the mixed column is the composition of the reactor outlet product obtained by backmixing automatically calculated by the system and used as the composition correction target.

[0179] Table 5 Reactor temperature drop and pressure drop correction targets

[0180] project unit Numeric A temperature drop ℃ 78 A reverse voltage drop kPa 2.3 Second anti-temperature drop ℃ 54 Second reverse voltage drop kPa 3

[0181] Table 6 Correction weights

[0182] project Weight Factor Ethylbenzene 0.066539 Styrene 0.066539 Toluene 0.066539 benzene 0.032379 water 0.000032 hydrogen 0.000032 Ethylene 0.03242 Methane 0.03242 Carbon monoxide 0.032379 carbon dioxide 0.032379 Cumene 0.03242 α-Methylstyrene 0.03242 o-Xylene 0.000032 Meta-Xylene 0.000032 p-Xylene 0.000032 o-Diethylbenzene 0.000032 m-Diethylbenzene 0.000032 p-Diethylbenzene 0.000032 Non Fang 0.000032 1 Anti-temperature drop 0.221798 1Reverse voltage drop 0.221798 2 Anti-temperature drop 0.064841 2 Reverse voltage drop 0.064841

[0183] It should be noted that the correction weights in Table 6 are the results after automatic normalization by the system.

[0184] Table 7 Convergence parameters

[0185]

[0186]

[0187] The reaction model was calibrated by running the gradient descent method to obtain the correction coefficients shown in Table 8. The comparison between the simulated values ​​after correction and the simulated values ​​and the target values ​​before correction is shown in Table 9.

[0188] Table 8 Correction coefficients

[0189] Calibration items Initial value Terminal Value <![CDATA[r1]]> 1 1.3 <![CDATA[r2]]> 1 0.65 <![CDATA[r3]]> 1 0.56 <![CDATA[r4]]> 1 0.6 <![CDATA[r5]]> 1 0.8 <![CDATA[r6]]> 1 0.3 <![CDATA[r7]]> 1 0.41 <![CDATA[r8]]> 1 0.615 <![CDATA[r9]]> 1 0.915 First reactor temperature drop 1 0.913 First reactor pressure drop 1 1.121 Second reactor temperature drop 1 0.923 Second reactor pressure drop 1 1.321

[0190] Table 9 Comparison of factory target values ​​before and after correction

[0191]

[0192]

[0193] It can be seen from Table 9 that after correction, the main by-products such as toluene, benzene, methane, carbon monoxide, carbon dioxide and the reactor temperature drop and pressure drop are closer to the factory values.

[0194] The present invention also proposes a simulation and correction system for the reaction of ethylbenzene dehydrogenation to styrene according to the above method, comprising:

[0195] Reaction kinetics module, used to calculate the reaction rates of each reaction in the reactor;

[0196] Reaction material balance module, used to calculate the flow and composition of each reaction raw material and reaction product in the reactor;

[0197] Reaction energy balance module, used to calculate the temperature change and reaction heat in the reactor;

[0198] Reaction momentum balance module, used to calculate the pressure change in the reactor;

[0199] A correction calculation module is used to correct the reaction model constants and simulate the content of each component in the reaction product, the temperature of the reaction product, and the pressure of the reaction product after correction;

[0200] The reaction of the reactor includes:

[0201] Reaction 1: Ethylbenzene reacts to produce styrene and hydrogen;

[0202] Reaction 2: Ethylbenzene reacts to produce benzene and ethylene;

[0203] Reaction 3: Ethylbenzene and hydrogen react to form toluene and methane;

[0204] Reaction 4: Styrene and hydrogen react to produce benzene and ethylene;

[0205] Reaction 5: Styrene and hydrogen react to produce toluene and methane;

[0206] Reaction 6: Methane and water react to produce carbon monoxide and hydrogen;

[0207] Reaction 7: Carbon monoxide reacts with water to produce carbon dioxide and hydrogen;

[0208] Reaction 8: Ethylene and water react to produce carbon monoxide and hydrogen;

[0209] Reaction 9: Isopropylbenzene reacts to produce α-methylstyrene and hydrogen.

[0210] The heat exchange calculation module, the reactor includes a first reactor and a second reactor connected in series; a heat exchanger is provided on the second reactor, and the heat exchange calculation module is used to calculate the heat exchange temperature and heat exchange load of the heat exchanger.

[0211] In addition, it should be noted that the above are only preferred embodiments of the present invention and are not limited to the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A simulation and correction method for the dehydrogenation of ethylbenzene to produce styrene, characterized in that: The following steps are involved: Step 1) Establish the reaction kinetics equation of the following chemical reaction: Reaction 1: Ethylbenzene reacts to produce styrene and hydrogen; Reaction 2: Ethylbenzene reacts to produce benzene and ethylene; Reaction 3: Ethylbenzene and hydrogen react to form toluene and methane; Reaction 4: Styrene and hydrogen react to produce benzene and ethylene; Reaction 5: Styrene and hydrogen react to produce toluene and methane; Reaction 6: Methane and water react to produce carbon monoxide and hydrogen; Reaction 7: Carbon monoxide reacts with water to produce carbon dioxide and hydrogen; Reaction 8: Ethylene and water react to produce carbon monoxide and hydrogen; Reaction 9: Cumene reacts to produce α-methylstyrene and hydrogen; Step 2) establishing a material balance equation for each reactant and reaction product according to the reaction kinetics equation in step 1); Step 3) establishing an energy balance equation according to the material balance equation in step 2); Step 4) establishing a momentum balance equation according to the material balance equation in step 2 and the energy balance equation in step 3); Step 5) establishing an ethylbenzene dehydrogenation radial adiabatic reactor model using the reaction kinetics equation, material balance equation, energy balance equation and momentum balance equation, and simulating and calculating the content of each component in the reaction product output by the reactor, the temperature of the reaction product, and the pressure of the reaction product according to the content of each component in the raw material fed into the reactor, the temperature of the raw material, the pressure of the reaction product, and the structural parameters of the reactor; Step six) establishing a correction objective function, and then establishing a batch gradient descent algorithm required for the correction calculation, with the goal of minimizing the correction objective function, correcting the constants in the radial adiabatic reactor model for ethylbenzene dehydrogenation; simulating the corrected ethylbenzene dehydrogenation radial adiabatic reactor model to obtain the corrected content of each component in the reaction product, the reaction product temperature, and the reaction product pressure; the correction objective function is: Where n is the number of system components, m is the number of reactors, x cal is the simulated value of the component molar fraction, x real is the target value of component molar fraction, w i is the component weight factor, w j is the reactor temperature drop / pressure drop weight factor, ΔT cal is the simulated value of reactor temperature drop, ΔT real is the target value of reactor temperature drop, ΔP cal is the simulated value of reactor pressure drop, ΔP real is the target value of reactor pressure drop.

2. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 1, characterized in that: The reaction kinetics equation includes: K P =C1×exp(C2-C3 / T-C4×log(T)) Among them, r1~r9 are the reaction rates of reactions 1~9, Kp is the main reaction equilibrium constant, C1~C4 are the correlation constants, and k i is the reaction rate constant, k i0 is the reaction rate pre-exponential factor, E a is the activation energy, b j is the adsorption constant, b j0 is the adsorption pre-exponential factor, Q j is the adsorption activation energy; P ST is the partial pressure of styrene, P EB is the partial pressure of ethylbenzene, P H2O is the water partial pressure, P H2 is the hydrogen partial pressure, P C2H4 is the ethylene partial pressure, P CH4 is the methane partial pressure, P CO is the partial pressure of carbon monoxide, P ISB is the partial pressure of cumene.

3. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 1, characterized in that: The material balance equation is: The flow rate change rate of impurity components that do not participate in the reaction is 0: Where F is the molar flow rate of the component, r1~r9 are the reaction rates of reactions 1~9, r is the bed radius, L is the bed height, ρ is the catalyst packing density, and F Other F stands for impurities; ST is the molar flow rate of styrene, F TL is the molar flow rate of toluene, F EB is the molar flow rate of ethylbenzene, F BZ is the molar flow rate of benzene, F H2O is the water molar flow rate, F H2 is the hydrogen molar flow rate, F C2H4 is the molar flow rate of ethylene, F CH4 is the molar flow rate of methane, F CO is the molar flow rate of carbon monoxide, F CO2 is the molar flow rate of carbon dioxide, F ISB is the molar flow rate of cumene, F AMS is the molar flow rate of α-methylstyrene, F other It is an impurity that does not participate in the reaction.

4. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 1, characterized in that: The energy balance equation is: Where ΔH j is the molar reaction enthalpy change, j from 1 to 9 represents reactions 1 to 9 in sequence; C pi is the molar heat capacity at constant pressure, i from 1 to 19 represents the components ethylbenzene, styrene, toluene, benzene, water, hydrogen, ethylene, methane, carbon monoxide, carbon dioxide, isopropylbenzene, α-methylstyrene, o-xylene, m-xylene, p-xylene, o-diethylbenzene, m-diethylbenzene, p-diethylbenzene, and non-aromatics.

5. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 1, characterized in that: The momentum balance equation is: Where ε is the bed porosity, d p is the equivalent diameter of the specific surface area of ​​the catalyst particles, μ is the dynamic viscosity of the mixed gas, and u is the superficial velocity of the gas.

6. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 1, characterized in that: The reactor in the step five) includes a first reactor and a second reactor connected in series, and the reaction kinetics equation, material balance equation, energy balance equation and momentum balance equation are used to sequentially calculate the reaction product temperature, reaction product pressure and reaction product composition of the first reactor and the second reactor.

7. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 6, characterized in that: The second reactor is provided with a heat exchanger, and the heat exchange temperature and heat exchange load of the heat exchanger are calculated using the following formula: Where T in,steam is the steam inlet temperature, T out,steam is the steam outlet temperature, F steam is the steam molar flow rate, C p,steam is the molar heat capacity of steam at constant pressure, T in,Material is the inlet temperature of the reaction materials, T out,Material is the outlet temperature of the reaction materials, F Material is the molar flow rate of the reaction materials, C p,Material is the molar heat capacity of the reaction materials at constant pressure.

8. The simulation and correction method for the reaction of ethylbenzene dehydrogenation to styrene according to claim 7, characterized in that: The batch gradient descent algorithm required for establishing the correction calculation is used to correct the model constants of the radial adiabatic reactor for ethylbenzene dehydrogenation with the goal of minimizing the correction objective function. The calculation steps are as follows: Step 1: Input the parameters required for calibration to form the initial task; Step 2: Add the initial task to the task queue and input the batch gradient descent method convergence parameters: minimum learning rate, minimum branch rate, error threshold eps, number of single-round iterations k, and initial learning rate α; Step 3: If the task queue is not empty, select the task with the smallest error f in the queue for optimization. If the queue is empty, the iteration fails and the program ends. Step 4: Given the initial point x (0) ∈R n , calculate the error function f; Step 5: Calculate the (k) The gradient at Step 6: Let d (k) =-g k ; Step 7. From x (k) Departure in direction d (k) Do a one-dimensional search, let x (k+1) =x (k) +α·d (k) ; Step 8. Check whether the convergence criterion is satisfied. If f < eps, stop the iteration, and the approximate minimum point is x (k) , otherwise go to 9; Step 9: If f is less than the branch threshold, create a new task with the same number of iterations and a learning rate halved, and add it to the task queue; Step 10, let k = k + 1, if k is equal to the written single-round iteration number, the iteration stops, indicating that this round of iteration fails, and returns to step 3; otherwise, returns to step 5; Preferably, when running the batch gradient descent algorithm, the solution parameters and correction calculation parameters of the two-stage ethylbenzene dehydrogenation radial adiabatic reactor model with an intermediate heat exchanger are input; The parameters to be solved for the two-stage ethylbenzene dehydrogenation radial adiabatic reactor model with an intermediate heat exchanger include: initial values ​​of kinetic parameters of each reaction; structural parameters of the styrene unit reactor and catalyst specifications, including the inner diameter, outer diameter, height, void ratio, catalyst bulk density, and catalyst equivalent diameter of the catalyst bed of the first reactor, and the inner diameter, outer diameter, height, void ratio, catalyst bulk density, and catalyst equivalent diameter of the catalyst bed of the second reactor; industrial feed flow, temperature, pressure, and composition; temperature, pressure, and flow of steam in the intermediate heat exchanger; and temperature and pressure of the raw material after exchange through the intermediate heat exchanger; The correction calculation parameters include: the actual production flow, temperature, pressure and composition of the dehydrogenated product in the same period as the feed data; the reactor inlet and outlet temperature and pressure; and the correction weight factors of the temperature drop and pressure drop of each component and each reactor.

9. A simulation and calibration system for the dehydrogenation of ethylbenzene to produce styrene, characterized in that: include: Reaction kinetics module, used to calculate the reaction rates of each reaction in the reactor; Reaction material balance module, used to calculate the flow and composition of each reaction raw material and reaction product in the reactor; Reaction energy balance module, used to calculate the temperature change and reaction heat in the reactor; Reaction momentum balance module, used to calculate the pressure change in the reactor; A correction calculation module is used to correct the reaction model constants and simulate the content of each component in the reaction product, the temperature of the reaction product, and the pressure of the reaction product after correction; The reaction of the reactor includes: Reaction 1: Ethylbenzene reacts to produce styrene and hydrogen; Reaction 2: Ethylbenzene reacts to produce benzene and ethylene; Reaction 3: Ethylbenzene and hydrogen react to form toluene and methane; Reaction 4: Styrene and hydrogen react to produce benzene and ethylene; Reaction 5: Styrene and hydrogen react to produce toluene and methane; Reaction 6: Methane and water react to produce carbon monoxide and hydrogen; Reaction 7: Carbon monoxide reacts with water to produce carbon dioxide and hydrogen; Reaction 8: Ethylene and water react to produce carbon monoxide and hydrogen; Reaction 9: Isopropylbenzene reacts to produce α-methylstyrene and hydrogen.

10. The simulation and correction system for the dehydrogenation of ethylbenzene to produce styrene according to claim 9, characterized in that: The reactor comprises a first reactor and a second reactor connected in series; Preferably, a heat exchanger is provided on the second reactor, and the system further comprises: The heat exchange calculation module is used to calculate the heat exchange temperature and heat exchange load of the heat exchanger.

Images