Method for establishing series reactor prediction model based on two reaction networks

Abstract

According to the method for establishing the series reactor prediction model based on the two reaction networks, a series reactor comprises a riser reactor and a fluidized bed reactor, raw materials sequentially pass through the riser reactor and the fluidized bed reactor to react to generate a product, and the product flows out; the method comprises the following steps: S1, according to a lumped kinetics principle and a reaction mechanism, performing lumped division on raw materials and products, and establishing a reaction network; s2, establishing a kinetic model based on the reaction network; s3, establishing a prediction model according to the dynamic model; the reaction network comprises a first reaction network and a second reaction network, the first reaction network corresponding to the riser reactor and the second reaction network corresponding to the fluidized bed reactor are established, and then the prediction model of the riser reactor and the prediction model of the fluidized bed reactor are established; and the simulation precision of the prediction model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of catalytic cracking, and specifically relates to a method for establishing a prediction model of a series reactor based on two reaction networks. Background Art

[0002] The riser reactor has the advantages of a relatively high operating linear velocity, good conveying performance, low backmixing degree, high reaction efficiency, and easy control of target products. The fluidized bed reactor has the advantages of high mass and heat transfer efficiency. The riser and fluidized bed series reactor combines the advantages of both, that is, it retains the characteristics of high reaction intensity of the riser reactor and utilizes the advantage of the reaction depth of the fluidized bed reactor, achieving high raw material conversion rate and ensuring high selectivity of the target product. To give full play to the advantages of the riser reactor and the fluidized bed reactor, a prediction model is established for the series reactor to guide the complementary advantages of the riser and fluidized bed reactors.

[0003] In the prior art, the following are several relatively representative kinetic models of the riser and fluidized bed series reactors:

[0004] "New Catalytic Cracking MIP Lumping Reaction Kinetics Model" proposed by Duan Liangwei, Sun Peng, Weng Huixin, etc. in 2012. The riser reactor and the fluidized bed reactor adopt the same lumping division and reaction network. The raw materials and products are divided into 10 lumps, namely raw material saturates, aromatics, resins and asphaltenes, diesel, saturated hydrocarbons in gasoline, olefins, aromatics, liquefied gas, gas, and coke. Based on this, a reaction device model is established;

[0005] "Heavy Oil Catalytic Cracking MIP Process Lumping Kinetics Model" proposed by Jiang Hongbo, Zhong Guijiang, Ning Hui, etc. in 2010. The riser reactor and the fluidized bed reactor adopt the same lumping division and reaction network. The raw materials and products are divided into 12 lumps, namely alkyl carbon, naphthene carbon, aromatic carbon in heavy oil, alkyl carbon, naphthene carbon, aromatic carbon in diesel, saturated hydrocarbons, olefins, aromatics, propylene, residual gas, and coke. Based on this, a reaction device model is established;

[0006] "Numerical Simulation of the Reaction Process of MIP Riser Based on a Multi-Scale Model" proposed by Lu Bona, Cheng Congli, Lu Weimin, etc. in 2013. The riser reactor and the fluidized bed reactor adopt the same lumping division and reaction network. The raw materials and products are divided into 12 lumps, namely alkyl carbon, naphthene carbon, side chains of aromatic hydrocarbons, benzene rings of aromatic hydrocarbons in heavy oil, alkyl carbon, naphthene carbon, side chains of aromatic hydrocarbons, benzene rings of aromatic hydrocarbons in diesel, gasoline, dry gas, liquefied gas, and coke. Based on this, a reaction device model is established;

[0007] The patent applied for by the present applicant in 2021 provides a method and device for evaluating the yield of catalytic cracking reaction products (Patent No.: CN116072233A). The catalytic cracking reaction is divided into 11 feed lumps and 6 product lumps. A reaction network and a catalytic cracking kinetic model are established according to the divided lumps. Then, the content of each feed lump is predicted based on the feed properties, and the yield of each product lump is predicted according to the catalytic cracking kinetic model.

[0008] However, the reaction time in the riser reactor is short, and only cracking reaction and condensation reaction occur in this reactor; the reaction time in the fluidized bed reactor is long, and cracking, hydrogen transfer, aromatization and condensation reactions can occur in this reactor. The above lump division is difficult to simulate all the reactions in the riser reactor and the fluidized bed reactor, which in turn leads to inaccurate prediction of the product yield by the prediction model. Summary of the Invention

[0009] In order to solve the problem in the prior art that the established prediction model is not precise enough due to inaccurate lump division, the present invention provides a method for establishing a prediction model for a series reactor based on two reaction networks, establishing a first reaction network corresponding to the riser reactor and a second reaction network corresponding to the fluidized bed reactor, and then establishing a prediction model for the riser reactor and a prediction model for the fluidized bed reactor, improving the simulation accuracy of the prediction model.

[0010] To achieve the above object, the specific solution adopted by the present invention is: a method for establishing a prediction model for a series reactor based on two reaction networks, the series reactor includes a riser reactor and a fluidized bed reactor, and the feed flows through the riser reactor and the fluidized bed reactor in sequence and then reacts to generate products and flows out;

[0011] S1, according to the lump kinetic principle and reaction mechanism, perform lump division on the feed and products and establish a reaction network;

[0012] S2, establish a kinetic model based on the reaction network;

[0013] S3, establish a prediction model according to the kinetic model;

[0014] The reaction network includes a first reaction network and a second reaction network. The lump division of the first reaction network is as follows: according to the distillation range and the reaction mechanism of the riser reactor, the feed in the riser reactor is divided into light oil layer, wax oil layer and residue oil layer, and further divided into light oil layer non-aromatic hydrocarbon lump, light oil layer aromatic hydrocarbon lump, wax oil layer non-aromatic hydrocarbon lump, wax oil layer aromatic hydrocarbon lump, residue oil layer non-aromatic hydrocarbon and residue oil layer aromatic hydrocarbon. The products of the riser reactor are divided into dry gas lump, liquefied gas lump, gasoline saturated hydrocarbon lump, gasoline olefin lump, gasoline aromatic hydrocarbon lump and coke lump;

[0015] The lumping of the second reaction network is as follows. The raw materials in the fluidized bed reactor are the products in the riser reactor. According to the distillation range and the reaction mechanism of the fluidized bed reactor, the raw materials in the fluidized bed reactor are divided into a heavy oil layer, a light oil layer, and a gasoline layer, and further divided into a heavy oil non-aromatic lumping, a heavy oil aromatic lumping, a light oil non-aromatic lumping, a light oil aromatic lumping, a gasoline saturated hydrocarbon lumping, a gasoline olefin lumping, and a gasoline aromatic lumping. The products in the fluidized bed reactor are divided into a dry gas lumping, a liquefied gas lumping, and a coke lumping.

[0016] As an optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The first reaction network established is as follows:

[0017] The heavy oil non-aromatic lumping reacts with the wax oil non-aromatic lumping, the light oil non-aromatic lumping, the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively;

[0018] The wax oil non-aromatic lumping reacts with the light oil non-aromatic lumping, the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively;

[0019] The gasoline non-aromatic lumping reacts with the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively;

[0020] The heavy oil aromatic lumping reacts with the wax oil aromatic lumping, the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively;

[0021] The wax oil aromatic lumping reacts with the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively;

[0022] The light oil aromatic lumping reacts with the dry gas lumping, the liquefied gas lumping, the gasoline aromatic lumping, and the coke lumping respectively.

[0023] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The second reaction network established is as follows:

[0024] The heavy oil non-aromatic lumping reacts with the light oil non-aromatic lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively;

[0025] The light oil non-aromatic lumping reacts with the gasoline saturated hydrocarbon lumping, the gasoline olefin lumping, the gasoline aromatic lumping, the dry gas lumping, and the liquefied gas lumping respectively;

[0026] The heavy oil aromatic lumping reacts with the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively;

[0027] The light oil aromatic lumping reacts with the gasoline aromatic lumping, the dry gas lumping, the liquefied gas lumping, and the coke lumping respectively;

[0028] The gasoline saturate lumps establish reactions with the dry gas lumps and the liquefied gas lumps respectively;

[0029] The gasoline olefin lumps establish reactions with the gasoline saturate lumps, the dry gas lumps, the liquefied gas lumps and the gasoline aromatic lumps respectively;

[0030] The gasoline aromatic lumps establish reactions with the dry gas lumps, the liquefied gas lumps and the coke lumps respectively.

[0031] As another optimization scheme of the above method for establishing a predictive model of a series reactor based on two reaction networks: The predictive model in S3 includes a riser reactor predictive model and a fluidized bed predictive model. The riser predictive model is:

[0032]

[0033] Where, X Riser = x Riser / H Riser is the dimensionless relative distance at the x Riser section in the bed layer, H Riser represents the height of the riser reactor, y i is the mass fraction of each component, x Riser represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the i-th lump in the riser reactor, Y Riser is the mass fraction vector of each lump component in the riser reactor, S WH is the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, P is the pressure in the riser reactor (Pa);

[0034] The fluidized bed predictive model is:

[0035]

[0036] Where, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed section in the fluidized bed layer, x bed is the distance from the fluidized bed inlet into the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the catalyst fixed carbon in the fluidized bed reactor, k W is the catalyst fixed carbon influence constant in the fluidized bed (cm 3 / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), S WH is the true weight hourly space velocity, yi is the mass fraction of each component, M i is the average relative molecular weight of the i-th lumping in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, and Y Bed is the mass fraction vector of each lumping component in the fluidized bed reactor.

[0037] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The kinetic model in S2 includes a riser reactor kinetic model and a fluidized bed reactor kinetic model.

[0038] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The riser kinetic model is derived from the riser continuity equation and the reaction rate:

[0039] The riser continuity equation is:

[0040]

[0041] In the first reaction network, the reaction rate of the j-th reaction is:

[0042]

[0043] The riser kinetic model is:

[0044]

[0045] Among them, ρ represents the density of the oil-gas mixture in the riser reactor (g / cm 3 ), t represents the reaction time, G v represents the mass flow rate of the oil-gas cross-section in the riser reactor (g / (cm 2 ·h)), x Riser represents the distance from the riser inlet into the reactor, a i is the concentration of the i-th lumping (molesi / g gas), R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the riser reactor, v i,j is the stoichiometric coefficient of the i-th lumping in the j-th reaction, r j represents the rate of the j-th reaction, P is the pressure in the riser reactor (Pa), k j is the reaction rate constant of the j-th reaction (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the riser reactor (g / cm 3), where ε is the void fraction, P is the system pressure (Pa) of the riser reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature (K) in the riser reactor, and a = [a1,..., a 12 T is the lumped component concentration vector.

[0046] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The fluidized bed kinetic model is derived from the continuity equation and reaction rate of the fluidized bed reactor:

[0047] The continuity equation of the fluidized bed reactor is:

[0048]

[0049] In the second reaction network, the reaction rate of the jth reaction is:

[0050]

[0051] The kinetic model of the fluidized bed reactor is:

[0052]

[0053] where ρ represents the density of the oil-gas mixture (g / cm 3 ) in the fluidized bed reactor, t represents the reaction time, G v represents the cross-sectional mass flow rate of the oil-gas (g / (em 2 ·h)) in the fluidized bed reactor, x Riser represents the distance from the fluidized bed inlet into the reactor, a i is the concentration of the ith lumped component (molesi / g gas), R i represents the reaction rate of the ith lumped component, n r is the number of reactions in the fluidized bed reactor, v i，j is the stoichiometric coefficient of the ith lumped component in reaction j, r j represents the rate of the jth reaction, P is the pressure (Pa) in the fluidized bed reactor, k j is the reaction rate constant of reaction j (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the fluidized bed reactor (g / cm 3 ), ε is the void fraction, P is the system pressure (Pa) of the fluidized bed reactor, R is the gas constant (8,314 J / (mol·K)), T is the system temperature (K) in the fluidized bed reactor, and a = [a1,..., a 10 T is the lumped component concentration vector.​​

[0054] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The prediction model in S3 includes a riser reactor prediction model and a fluidized bed reactor prediction model. Both the riser reactor prediction model and the fluidized bed reactor prediction model introduce a catalyst carbon content function to characterize its influence on the yield. The catalyst carbon content function in the riser reactor prediction model is:

[0055] θ(C Riser ) = exp(-k W ·C Riser )

[0056] Where C Riser is the catalyst carbon content (w%) in the riser reactor, and k W is the catalyst carbon content influence constant;

[0057]

[0058] Where y CK is the fraction of coke in the riser reactor, and R Co is the catalyst-to-oil ratio in the riser reactor;

[0059] The catalyst carbon content function in the fluidized bed reactor prediction model is:

[0060] θ(C Bed ) = exp(-k W ·C Bed )

[0061] Where C Bed is the catalyst carbon content (w%) in the fluidized bed reactor, and k W is the catalyst carbon content influence constant;

[0062]

[0063] Where R CO represents the catalyst-to-oil ratio in the fluidized bed reactor, y CK，Riser，out is the fraction of coke at the outlet of the riser reactor, and β is the influence degree of reactor structure, reaction conditions, and feedstock oil properties on the catalyst carbon content.

[0064] As another optimization scheme of the above method for establishing a prediction model of a series reactor based on two reaction networks: The prediction model of the riser reactor is:

[0065]

[0066] Where X Riser = x Riser / HRiser The dimensionless relative distance, H, at the x cross-section in the bed layer Riser represents the height of the riser reactor, y Riser is the mass fraction of each component, x i represents the distance from the riser inlet into the reactor, M Riser is the average relative molecular weight of the i-th lumping in the riser reactor, Y i is the vector of mass fractions of lumped components in the riser reactor, S Riser is the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, exp(-k WH ·C W ) is the catalyst carbon function in the riser reactor, and P is the pressure (Pa) in the riser reactor. Riser

[0067] As another optimization scheme of the method for establishing a prediction model of a series reactor based on two reaction networks: the prediction model of the fluidized bed reactor is:

[0068]

[0069] where X Bed = x Bed / H Bed is the dimensionless relative distance at the x cross-section in the fluidized bed layer bed x is the distance from the fluidized bed inlet into the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the catalyst carbon in the fluidized bed reactor, k bed is the catalyst carbon influence constant in the fluidized bed (cm W / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature (K) in the fluidized bed reactor, S 3 is the true weight hourly space velocity, y WH is the mass fraction of each component, M i is the average relative molecular weight of the i-th lumping in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, Y i is the vector of mass fractions of lumped components in the fluidized bed reactor. Bed

[0070] An apparatus for establishing a prediction model of a series reactor based on two reaction networks, the establishing apparatus includes a modeling module that models using the above riser and fluidized bed series catalytic cracking reaction models.

[0071] ​​An electronic device includes a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the above prediction model can be run.

[0072] A readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the above method are run.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] 1. The present invention provides a method for establishing a prediction model of a series reactor based on two reaction networks. Different lumps are divided corresponding to the riser reactor and the fluidized bed reactor, and a first reaction network and a second reaction network are established, significantly improving the simulation accuracy of the model. At the same time, the prediction model provides guidance for the optimal operation of the riser and fluidized bed series reactor under coupling conditions, gives play to the respective advantages of the riser reactor and the fluidized bed reactor, and realizes the complementary advantages of the two reactors of the riser and the fluidized bed.

[0075] 2. A catalyst coke content function is introduced into the prediction model to characterize the influence of the change of catalyst activity on the prediction model. Among them, the introduction of the catalyst coke content function in the prediction model of the riser reactor shows the change of catalyst activity in its axial direction. The catalyst enters the fluidized bed reactor after reacting in the riser reactor and further accumulates carbon in the fluidized bed reactor. The degree of carbon accumulation is affected by the reactor structure, reaction conditions and feedstock oil properties. Therefore, β is introduced into the catalyst coke content function in the prediction model of the fluidized bed reactor to make the prediction results of the yields of each product lump more accurate. Specific embodiments

[0076] The technical solutions of the present invention will be further elaborated in detail below in conjunction with specific embodiments. For parts not detailedly recorded and disclosed in the following embodiments of the present invention, they should all be understood as the prior art known or should be known to those skilled in the art.

[0077] Embodiment 1

[0078] A method for establishing a prediction model of a series reactor based on two reaction networks. The series reactor includes a riser reactor and a fluidized bed reactor. The raw material passes through the riser reactor and the fluidized bed reactor in sequence, reacts and then the product flows out.

[0079] S1. According to the principles of lumped kinetics and reaction mechanism, the raw materials and products are lumped and a reaction network is established. The reactions in the riser reactor and the fluidized bed reactor have different characteristics. Therefore, the reaction network includes a first reaction network and a second reaction network. The first reaction network corresponds to the reactions in the riser reactor, and the second reaction network corresponds to the reactions in the fluidized bed reactor. Different lumps are divided corresponding to the riser reactor and the fluidized bed reactor, and the first reaction network and the second reaction network are established, significantly improving the simulation accuracy of the model.

[0080] The lumping of the first reaction network is as follows: According to the distillation range and the reaction mechanism of the riser reactor, the raw materials in the riser reactor are divided into a light oil layer (200°C - 350°C), a wax oil layer (350°C - 500°C), and a residue oil layer (>500°C). The reaction conditions in the riser reactor are harsh, mainly undergoing high-intensity cracking reactions and condensation reactions, separating the aromatics and non-aromatics in the raw materials, and dividing them into a light oil layer non-aromatic lumps, a light oil layer aromatic lumps, a wax oil layer non-aromatic lumps, a wax oil layer aromatic lumps, a residue oil layer non-aromatic, and a residue oil layer aromatic; only products with a temperature lower than 200°C and coke are considered in the riser reactor. Specifically, the products of the riser reactor are divided into dry gas lumps, liquefied gas lumps, gasoline saturated hydrocarbon lumps, gasoline olefin lumps, gasoline aromatic lumps, and coke lumps.

[0081] The reactions in the riser reactor have the following characteristics: The reaction time is short, and only cracking reactions and condensation reactions occur in this reactor; the reaction temperature is relatively high, and all reactions of large molecules cracking into small molecules are possible; dry gas, liquefied gas, gasoline, and coke are the final products, and wax oil non-aromatic, wax oil aromatic, light oil non-aromatic, and light oil aromatic are both reactants and products; non-aromatics and aromatics do not generate each other; condensation reactions can occur in the residue oil layer, wax oil layer, and light oil layer aromatic to generate coke. Based on this, the established first reaction network is as follows:

[0082] The residue oil layer non-aromatic lumps establish reactions with the wax oil layer non-aromatic lumps, the light oil layer non-aromatic lumps, the dry gas lumps, the liquefied gas lumps, the gasoline saturated hydrocarbon lumps, and the gasoline olefin lumps respectively;

[0083] The wax oil layer non-aromatic lumps establish reactions with the light oil layer non-aromatic lumps, the dry gas lumps, the liquefied gas lumps, the gasoline saturated hydrocarbon lumps, and the gasoline olefin lumps respectively;

[0084] The gasoline non-aromatic lumps establish reactions with the dry gas lumps, the liquefied gas lumps, the gasoline saturated hydrocarbon lumps, and the gasoline olefin lumps respectively;

[0085] The residue oil layer aromatic lumps establish reactions with the wax oil layer aromatic lumps, the light oil layer aromatic lumps, the gasoline aromatic lumps, and the coke lumps respectively;

[0086] The lumped wax oil aromatics react with the lumped light oil aromatics, the lumped gasoline aromatics and the lumped coke respectively;

[0087] The lumped light oil aromatics react with the lumped dry gas, the lumped liquefied gas, the lumped gasoline aromatics and the lumped coke respectively.

[0088] After the raw material reacts in the riser reactor, its properties change significantly, and the raw material cannot be divided into a light oil layer, a wax oil layer and a residue oil layer according to the distillation range. After the raw material passes through the riser reactor, the product with an initial boiling point greater than 350 °C in the riser reactor product is slurry oil. The distillation range of this product overlaps with that of wax oil and residue oil, but its properties are significantly different from those of wax oil and residue oil. Therefore, the components with a distillation range greater than 350 °C are divided into the heavy oil layer in the fluidized bed reactor; the product with a distillation range of 200 °C to 350 °C in the riser reactor product is diesel. The distillation range of this product overlaps with that of light oil, but its properties are significantly different from those of light oil. In the fluidized bed reaction, the components with a distillation range of 200 °C to 350 °C are divided into the light oil layer. Therefore, the lumping of the second reaction network is that the raw material in the fluidized bed reactor is the product in the riser reactor. According to the distillation range and the reaction mechanism of the fluidized bed reactor, the raw material in the fluidized bed reactor is divided into a heavy oil layer, a light oil layer and a gasoline layer, and is divided into a heavy oil non-aromatic lumped component, a heavy oil aromatic lumped component, a light oil non-aromatic lumped component, a light oil aromatic lumped component, a gasoline saturated hydrocarbon lumped component, a gasoline olefin lumped component and a gasoline aromatic lumped component. The products in the fluidized bed reactor are divided into a dry gas lumped component, a liquefied gas lumped component and a coke lumped component.

[0089] The reaction in the fluidized bed reactor has the following characteristics: the reaction time is relatively long, and cracking reactions, hydrogen transfer reactions, aromatization reactions and condensation reactions may occur in this reactor; due to the relatively low reaction severity, the gas and coke yields are small. In this reactor, there is no reaction of large molecules cracking into small molecules and no reaction of small molecules condensing to form coke; dry gas, liquefied gas and coke are the final products, and heavy oil non-aromatics, heavy oil aromatics, gasoline saturated hydrocarbons, gasoline olefins and gasoline aromatics are both reactants and products; heavy oil non-aromatics can generate light oil aromatics through aromatization reactions; gasoline olefins can generate gasoline saturated hydrocarbons and gasoline aromatics through hydrogen transfer reactions. Based on this, the second reaction network established is as follows:

[0090] The heavy oil non-aromatic lumped component reacts with the light oil non-aromatic lumped component, the gasoline saturated hydrocarbon lumped component and the gasoline olefin lumped component respectively;

[0091] The light oil non-aromatic lumped component reacts with the gasoline saturated hydrocarbon lumped component, the gasoline olefin lumped component, the gasoline aromatic lumped component, the dry gas lumped component and the liquefied gas lumped component respectively;

[0092] The heavy oil aromatic lumped component reacts with the light oil aromatic lumped component, the gasoline aromatic lumped component and the coke lumped component respectively;

[0093] The light oil aromatic lumps react with the gasoline aromatic lumps, dry gas lumps, liquefied gas lumps and coke lumps respectively;

[0094] The gasoline saturated hydrocarbon lumps react with the dry gas lumps and liquefied gas lumps respectively;

[0095] The gasoline olefin lumps react with the gasoline saturated hydrocarbon lumps, dry gas lumps, liquefied gas lumps and gasoline aromatic lumps respectively;

[0096] The gasoline aromatic lumps react with the dry gas lumps, liquefied gas lumps and coke lumps respectively.

[0097] S2. Establish a kinetic model based on the reaction network.

[0098] The kinetic model of the riser reactor is obtained based on the continuity equation and reaction rate of the riser reactor:

[0099] The reaction rate of the chemical reaction in the riser reactor is:

[0100]

[0101] where k j is the reaction rate constant of reaction j in the riser reactor (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the riser reactor (g / cm 3 ), ε is the void fraction, P is the system pressure of the riser reactor (Pa), R is the gas constant (8.314 J / (mol·K)), and T is the system temperature in the riser reactor (K).

[0102]

[0103] where, A j is the pre-exponential factor, E j is the activation energy, R is the gas constant (8.314 J / (mol·K)), and T is the system temperature in the riser reactor (K).

[0104] For the i-th lumps in the riser reactor, its reaction rate equation is:

[0105]

[0106] where, n r is the number of reactions in the riser reactor, v i,j is the stoichiometric coefficient of the i-th lumps in reaction j, and r j represents the rate of the j-th reaction.

[0107] The continuity equation of the chemical reaction in the riser reactor is:

[0108]

[0109] In the first - reaction network, the reaction rate of the j - th reaction is:

[0110]

[0111] The riser - tube kinetic model is:

[0112]

[0113] where ρ represents the density of the oil - gas mixture in the riser - tube reactor (g / cm 3 ), t represents the reaction time, G v represents the mass flow rate of the oil - gas cross - section in the riser - tube reactor (g / (cm 2 ·h)), x Riser represents the distance from the riser - tube inlet into the reactor, a i is the concentration of the i - th lumping (molesi / g gas), R i represents the reaction rate of the i - th lumping, n r is the number of reactions in the riser - tube reactor, v i,j is the stoichiometric coefficient of the i - th lumping in the j - th reaction, r j represents the rate of the j - th reaction, P is the pressure in the riser - tube reactor (Pa), k j is the reaction rate constant of the j - th reaction in the riser - tube reactor (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the riser - tube reactor (g / cm 3 ), ε is the void fraction, P is the system pressure of the riser - tube reactor (Pa), R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the riser - tube reactor (K), a = [a1,...,a 12 T is the lumped - component concentration vector.

[0114] The kinetic model of the fluidized - bed reactor is obtained based on the continuity equation and reaction rate of the fluidized - bed reactor:

[0115] The reaction rate of the chemical reaction in the fluidized - bed reactor is:

[0116]

[0117] where k j is the reaction rate constant of the j - th reaction in the fluidized - bed reactor (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the fluidized - bed reactor (g / cm​3 ), where ε is the void fraction, P is the system pressure (Pa) of the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), and T is the system temperature (K) in the fluidized bed reactor.

[0118] The reaction rate constant of reaction j in the riser reactor is:

[0119]

[0120] Among them, A j is the pre-exponential factor, E j is the activation energy, R is the gas constant (8.314 J / (mol·K)), and T is the system temperature (K) in the fluidized bed reactor.

[0121] The continuity equation of the chemical reaction in the fluidized bed reactor is:

[0122]

[0123] In the second reaction network, the reaction rate of the jth reaction is:

[0124]

[0125] The kinetic model of the fluidized bed reactor is:

[0126]

[0127] Among them, ρ represents the density (g / cm 3 ) of the oil-gas mixture in the fluidized bed reactor, t represents the reaction time, G v represents the cross-sectional mass flow rate (g / (cm 2 ·h)) of the oil-gas in the fluidized bed reactor, x Riser represents the distance from the fluidized bed inlet into the reactor, a i is the concentration (molesi / g gas) of the i-th lumping, R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the fluidized bed reactor, v i,j is the stoichiometric coefficient of the i-th lumping in reaction j, r j represents the rate of the jth reaction, P is the pressure (Pa) in the fluidized bed reactor, k j is the reaction rate constant (cm 3 / (g·h)) of reaction j in the fluidized bed reactor, ρ c is the catalyst density (g / cm 3), ε is the void fraction, P is the system pressure (Pa) of the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature (K) in the fluidized bed reactor, a = [a1,..., a 10 T is the lumped component concentration vector.

[0128] S3. Establish a prediction model based on the kinetic model. The prediction model includes a riser reactor prediction model and a fluidized bed reactor prediction model.

[0129] The prediction model of the riser reactor is:

[0130]

[0131] where, X Riser = x Riser / H Riser is the dimensionless relative distance at the x Riser cross-section in the bed, H Riser represents the height of the riser reactor, y i is the mass fraction of each component, x Riser represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the i-th lumped in the riser reactor, Y Riser is the mass fraction vector of each lumped component in the riser reactor, S WH is the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, and P is the pressure (Pa) in the riser reactor.

[0132] The mass fraction vector of each lumped component in the riser reactor is:

[0133] Y Riser = [y RR , y RA , y MR , y MA , y LR , y LA , y GP , y GO , y GA , y GAS , y LPG , y CK

[0134] where, y RR is the mass fraction of the residue non-aromatic hydrocarbon lumped, y RA is the mass fraction of the residue aromatic hydrocarbon lumped, y MR is the mass fraction of the gas oil non-aromatic hydrocarbon lumped, y MA is the mass fraction of the gas oil aromatic hydrocarbon lumped, y LR ​​is the mass fraction of the light oil non-aromatic hydrocarbon lumps, y LA is the mass fraction of the light oil aromatic hydrocarbon lumps, y GP is the mass fraction of the gasoline saturated hydrocarbon lumps, y GO is the mass fraction of the gasoline olefin lumps, y GA is the mass fraction of the gasoline aromatic hydrocarbon lumps, y GAS is the mass fraction of the dry gas lumps, y LPG is the mass fraction of the liquefied gas lumps, y CK is the mass fraction of the coke lumps.

[0135] The prediction model of the fluidized bed reactor is:

[0136]

[0137] Among them, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed section in the fluidized bed layer, x bed is the distance from the fluidized bed inlet into the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the fixed carbon of the catalyst in the fluidized bed reactor, k W is the influence constant of the fixed carbon of the catalyst in the fluidized bed (cm 3 / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), S WH is the true weight hourly space velocity, y i is the mass fraction of each component, M i is the average relative molecular weight of the i-th lump in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, Y Bed is the mass fraction vector of each lump component in the fluidized bed reactor.

[0138] The mass fraction vector of each lump component in the fluidized bed reactor is:

[0139] Y Bed = [y HR , y HA , y LR , y LA , y GP , y GO , y GA , y GAS , y LPG , y CK

[0140] Among them, y HR ​is the mass fraction of the heavy oil non-aromatic lumps, y HA is the mass fraction of the heavy oil aromatic lumps, y LR is the mass fraction of the light oil non-aromatic lumps, y LA is the mass fraction of the light oil aromatic lumps, y GP is the mass fraction of the gasoline saturated hydrocarbon lumps, y GO is the mass fraction of the gasoline olefin lumps, y GA is the mass fraction of the gasoline aromatic lumps, y GAS is the mass fraction of the dry gas lumps, y LPG is the mass fraction of the liquefied gas lumps, y CK is the mass fraction of the coke lumps.

[0141] For the series reactors, the inlet feedstock composition of the riser reactor is as follows:

[0142]

[0143] Among them, is the mass fraction of the residue non-aromatic lumps at the riser inlet, is the mass fraction of the residue aromatic lumps at the riser inlet, is the mass fraction of the wax oil non-aromatic lumps at the riser inlet, is the mass fraction of the wax oil aromatic lumps at the riser inlet, is the mass fraction of the light oil non-aromatic lumps at the riser inlet, is the mass fraction of the light oil aromatic lumps at the riser inlet. The inlet feedstock composition of the fluidized bed reactor is the same as the product composition of the riser reactor.

[0144] The prediction model provides guidance for the optimal operation of the series reactors of the riser and the fluidized bed under coupled conditions, gives full play to the respective advantages of the riser reactor and the fluidized bed reactor, and realizes the complementary advantages of the two reactors of the riser and the fluidized bed.

[0145] The above is the basic implementation mode of the present invention, and further improvements, optimizations and limitations can be made on this basis to obtain the following embodiments:

[0146] Embodiment 2

[0147] This embodiment is an improved method for establishing a prediction model of series reactors based on two reaction networks on the basis of Embodiment 1. Its main structure is the same as that of Embodiment 1, and the improvement lies in:

[0148] The catalyst carbon deposition function is introduced into both the prediction model of the riser reactor and the prediction model of the fluidized bed reactor to characterize its influence on the yield. The catalyst carbon deposition function in the prediction model of the riser reactor is:

[0149] θ(CRiser ) = exp(-k W ·C Riser )

[0150] Among them, C Riser is the fixed carbon of the catalyst in the riser reactor (w%), and k W is the influence constant of the fixed carbon of the catalyst;

[0151]

[0152] Among them, y CK is the fraction of coke in the riser reactor, and R CO is the catalyst-to-oil ratio in the riser reactor;

[0153] The fixed carbon function of the catalyst in the fluidized bed reactor prediction model is:

[0154] θ(C Bed ) = exp(-k W ·C Bed )

[0155] Among them, C Bed is the fixed carbon of the catalyst in the fluidized bed reactor (w%), and k W is the influence constant of the fixed carbon of the catalyst;

[0156] Among them, R CO represents the catalyst-to-oil ratio in the fluidized bed reactor, y CK，Riser，out is the fraction of coke at the outlet of the riser reactor, and β is the influence degree of reactor structure, reaction conditions, and feedstock oil properties on the fixed carbon of the catalyst.

[0157] Then the establishment process of the prediction model of the riser reactor is:

[0158] The reaction rate of the chemical reaction in the riser reactor is:

[0159]

[0160] Among them, θ(C Riser ) is the influence function of the fixed carbon of the catalyst in the riser reactor on the reaction rate, k j is the reaction rate constant of reaction j (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the riser reactor (g / cm 3 ), ε is the porosity, P is the system pressure of the riser reactor (Pa), R is the gas constant (8.314 J / (mol·K)), and T is the system temperature in the riser reactor (K).

[0161] The reaction rate constant of reaction j in the riser reactor is:

[0162]

[0163] For the i-th lumping in the riser reactor, its reaction rate equation is:

[0164]

[0165] where, n r is the number of reactions in the riser reactor, v i，j is the stoichiometric coefficient of the i-th lumping in reaction j, and r j represents the rate of the j-th reaction.

[0166] The continuity equation of the chemical reaction in the riser reactor is:

[0167]

[0168] where, ρ represents the density of the oil-gas mixture (g / cm 3 ), t represents the reaction time, G v represents the mass flow rate of the oil-gas cross-section (g / (em 2 ·h)), x Riser represents the distance from the riser inlet into the reactor, a i is the concentration of the i-th lumping (molesi / g gas), R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the riser reactor, v i，j is the stoichiometric coefficient of the i-th lumping in reaction j, r j represents the rate of the j-th reaction, and P is the pressure in the riser reactor (Pa).

[0169] The kinetic model of the riser reactor is:

[0170]

[0171] where, X Riser = x Riser / H Riser is the dimensionless relative distance at the x Riser cross-section in the bed, H Riser represents the height of the riser reactor, x Riser represents the distance from the riser inlet into the reactor, a = [a1,..., a 12 T is the vector of lumped component concentrations, θ(C Riser ) represents the catalyst fixed carbon function in the riser reactor, S WH is the true weight hourly space velocity, a iis the total concentration of the i-th set (molesi / g gas), K is the reaction rate constant matrix in the riser reactor, and P is the pressure in the riser reactor (Pa).

[0172] Among them, θ(C Riser ) = exp(-k W ·C Riser ), C Riser is the fixed carbon of the catalyst in the riser reactor (w%), and k W is the fixed carbon influence constant of the catalyst.

[0173] The prediction model of the riser reactor is:

[0174]

[0175] Among them, X Riser = x Riser / H Riser is the dimensionless relative distance at the x Riser section in the bed, H Riser represents the height of the riser reactor, y i is the mass fraction of each component, x Riser represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the i-th set in the riser reactor, Y Riser is the mass fraction vector of each lumped component in the riser reactor, S WH is the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, exp(-k W ·C Riser ) is the catalyst fixed carbon function of the riser reactor, and P is the pressure in the riser reactor (Pa).

[0176] The prediction model of the fluidized bed reactor is:

[0177]

[0178] Among them, X Riser = x Riser / H Riser is the dimensionless relative distance at the x Riser section in the bed, H Riser represents the height of the riser reactor, y i is the mass fraction of each component, x Riser represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the i-th set in the riser reactor, Y Riser is the mass fraction vector of each lumped component in the riser reactor, S WHis the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, exp(-k W ·C Riser ) is the catalyst carbon function of the riser reactor, and P is the pressure (Pa) in the riser reactor.

[0179] The mass fraction vector of each lumped component in the riser reactor is:

[0180] Y Riser = [y RR , y RA , y MR , y MA , y LR , y LA , y GP , y GO , y GA , y GAS , y LPG , y CK

[0181] Among them, y RR is the mass fraction of the residue non-aromatic hydrocarbon lumped component, y RA is the mass fraction of the residue aromatic hydrocarbon lumped component, y MR is the mass fraction of the gas oil non-aromatic hydrocarbon lumped component, y MA is the mass fraction of the gas oil aromatic hydrocarbon lumped component, y LR is the mass fraction of the light oil non-aromatic hydrocarbon lumped component, y LA is the mass fraction of the light oil aromatic hydrocarbon lumped component, y GP is the mass fraction of the gasoline saturated hydrocarbon lumped component, y GO is the mass fraction of the gasoline olefin lumped component, y GA is the mass fraction of the gasoline aromatic hydrocarbon lumped component, y GAS is the mass fraction of the dry gas lumped component, y LPG is the mass fraction of the liquefied gas lumped component, y CK is the mass fraction of the coke lumped component.

[0182] The oil and gas flow rate in the riser reactor is high and the residence time is short, so backmixing can be ignored. Therefore, the riser reactor can be regarded as a plug flow reactor. The catalyst carbon content changes along the axial direction of the riser reactor, making the value of the catalyst carbon function change along the axial direction of the riser reactor. According to the prediction model, the fraction of products in the riser reactor can be accurately calculated.

[0183] The establishment process of the fluidized bed prediction model is as follows:

[0184] The reaction rate of the chemical reaction in the fluidized bed reactor is:

[0185] ​

[0186] Among them, θ(C Bed ) is the influence function of the fixed carbon of the catalyst in the fluidized bed reactor on the reaction rate, k j is the reaction rate constant of reaction j (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the fluidized bed reactor (g / cm 3 ), ε is the porosity, P is the system pressure of the fluidized bed reactor (Pa), R is the gas constant (8.314 J / (mol·K)), and T is the system temperature in the fluidized bed reactor (K).

[0187] The reaction rate constant of reaction j in the fluidized bed reactor is:

[0188]

[0189] Among them, A j is the pre-exponential factor, E j is the activation energy, R is the gas constant (8.314 J / (mol·K)), and T is the system temperature in the fluidized bed reactor (K).

[0190] The continuity equation of the chemical reaction in the fluidized bed reactor is:

[0191]

[0192] Among them, ρ represents the density of the oil-gas mixture (g / cm 3 ), t represents the reaction time, Gv represents the mass flow rate of the oil-gas cross-section (g / (cm 2 ·h)), x Bed represents the distance from the inlet of the fluidized bed into the reactor, a i is the concentration of the i-th lumping (molesi / g gas), R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the fluidized bed reactor, v i,j is the stoichiometric coefficient of the i-th lumping in reaction j, r j represents the rate of the j-th reaction, and P is the pressure in the fluidized bed reactor (Pa).

[0193] The kinetic model of the fluidized bed reactor is:

[0194]

[0195] Among them, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed section in the fluidized bed layer, x bedis the distance from the fluidized bed inlet to the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the fixed carbon of the catalyst in the fluidized bed reactor, a = [a1,..., a 10 T is the lumped component concentration vector, θ(C Bed ) represents the fixed carbon function of the catalyst in the fluidized bed reactor, S WH is the true weight hourly space velocity, a i is the concentration of the i-th lumped component (molesi / g gas), K is the reaction rate constant matrix in the fluidized bed reactor, and P is the pressure in the fluidized bed reactor (Pa).

[0196] Among them, θ(C Bed ) = exp(-k W ·C Bed ), C Bed is the fixed carbon of the catalyst in the fluidized bed reactor (w%), and k W is the fixed carbon influence constant of the catalyst.

[0197] The prediction model of the fluidized bed reactor is as follows:

[0198]

[0199] Among them, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed section in the fluidized bed layer, x bed is the distance from the fluidized bed inlet to the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the fixed carbon of the catalyst in the fluidized bed reactor, k W is the fixed carbon influence constant of the catalyst in the fluidized bed (cm 3 / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), S WH is the true weight hourly space velocity, y i is the mass fraction of each component, M i is the average relative molecular weight of the i-th lumped component in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, and Y Bed is the mass fraction vector of each lumped component in the fluidized bed reactor.

[0200] Y Bed = [y HR , y HA , y LR , y​LA , y GP , y GO , y GA , y GAS , y LPG , y CK

[0201] Among them, y HR is the mass fraction of the heavy oil non-aromatic lumps, y HA is the mass fraction of the heavy oil aromatic lumps, y LR is the mass fraction of the light oil non-aromatic lumps, y LA is the mass fraction of the light oil aromatic lumps, y GP is the mass fraction of the gasoline saturated hydrocarbon lumps, y GO is the mass fraction of the gasoline olefin lumps, y GA is the mass fraction of the gasoline aromatic lumps, y GAS is the mass fraction of the dry gas lumps, y LPG is the mass fraction of the liquefied gas lumps, y CK is the mass fraction of the coke lumps. In the fluidized bed reactor, the oil and gas move in a plug flow upward along its axis, that is, the catalyst moves in a completely mixed flow in the fluidized bed reactor, and the catalyst activity remains stable in the fluidized bed reactor. Therefore, C Bed is a constant value, but not equal to the fixed carbon of the catalyst at the outlet of the riser (C Riser，out ). The catalyst leaves the riser reactor and enters the fluidized bed reactor to continue catalytic cracking, and it further deposits carbon, and the degree of carbon deposition is affected by factors such as the reactor structure, reaction conditions, and properties of the feedstock oil.

[0202] For the series reactor, the feedstock composition at the inlet of the riser reactor is:

[0203]

[0204] Among them is the mass fraction of the heavy oil non-aromatic lumps at the inlet of the riser, is the mass fraction of the heavy oil aromatic lumps at the inlet of the riser, is the mass fraction of the wax oil non-aromatic lumps at the inlet of the riser, is the mass fraction of the wax oil aromatic lumps at the inlet of the riser, is the mass fraction of the light oil non-aromatic lumps at the inlet of the riser, is the mass fraction of the light oil aromatic lumps at the inlet of the riser. The feedstock composition at the inlet of the fluidized bed reactor is the same as the product composition of the riser reactor.

[0205] In this embodiment, the feedstock composition of the fluidized bed reactor is the same as the product composition at the outlet of the riser reactor.

[0206] Example 3​

[0207] A device for establishing a prediction model of a series reactor based on two reaction networks, the device includes a modeling module that uses the series catalytic cracking reaction model of a riser and a fluidized bed described in Example 1 for modeling.

[0208] Example 4

[0209] An electronic device includes a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned evaluation method can be executed.

[0210] In this embodiment, the electronic device includes a modeling device, a processor, a memory, a storage controller, a peripheral interface, an input and output unit, an audio unit, a display unit, and the like.

[0211] Specifically, the memory, storage controller, processor, peripheral interface, input / output unit, audio unit, and display unit are electrically connected to each other directly or indirectly to achieve data transmission or interaction. The modeling device includes at least one software function module that can be stored in the memory in the form of software or firmware or fixed in the operating system (OS) of the modeling device. The processor is used to execute the executable modules stored in the memory, including software function modules or computer programs.

[0212] The memory may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc. The memory is used to store programs, and the processor executes the corresponding program after receiving the execution instruction. The method executed by the server defined by the flow process involved in this application can be applied to the processor or implemented by the processor.

[0213] The processor can be an integrated circuit chip with signal processing capabilities. The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0214] The peripheral interface couples various input / output devices to the processor and the memory. The peripheral interface, the processor, and the memory controller may be implemented in a single chip or may be implemented in separate chips.

[0215] The input / output unit, the audio unit, and the display unit are all prior arts. For example, the input / output unit is used to provide an interface for users to input data to implement the interaction between the user and the server (or local terminal), and it can be a mouse, a keyboard, etc.; for example, the audio unit provides an audio interface for users, which may include one or more microphones, one or more speakers, and an audio circuit; for example, the display unit provides an interaction interface (such as a user operation interface) between the electronic device and the user or is used to display image data for the user to refer to.

[0216] Embodiment 5

[0217] A readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps of the above evaluation method are run.

[0218] In order to verify the effect of the present invention, the following experimental examples are carried out:

[0219] Experimental Example 1

[0220] According to the method of Embodiment 1, a prediction model of a riser reactor and a prediction model of a fluidized bed reactor are established. The parameters in the models are fitted according to the test data. The properties of the test feedstock oil are shown in Table 1, the product distribution is shown in Table 2, and the average relative error between the fitted values and the test values calculated through Embodiment 1 is shown in Table 3.

[0221] Experimental Example 2

[0222] According to the method of Embodiment 2, a prediction model of a riser reactor and a prediction model of a fluidized bed reactor are established. The parameters in the models are fitted according to the test data. The properties of the test feedstock oil are shown in Table 1, the product distribution is shown in Table 2, and the average relative error between the fitted values and the test values calculated through Embodiment 1 is shown in Table 3.

[0223] Comparative Example 1

[0224] This Comparative Example 1 is a reaction device model established according to the "New Catalytic Cracking MIP Lumping Reaction Kinetics Model" proposed by Duan Liangwei, Sun Peng, Weng Huixin, etc. in 2012. The parameters in the model are fitted according to the same test data as in the experimental examples. The comparison between the fitted values and the test values calculated is shown in Table 3 below.

[0225] Comparative Example 2

[0226] This Comparative Example 2 is a reaction device model established according to the "Lumping Kinetics Model of Heavy Oil Catalytic Cracking MIP Process" proposed by Jiang Hongbo, Zhong Guijiang, Ning Hui, etc. in 2010. The parameters in the model are fitted according to the same test data as in the experimental examples. The comparison between the fitted values and the test values calculated is shown in Table 3 below.

[0227] Comparative Example 3

[0228] Comparative Example 3 is a reaction device model established based on "Numerical Simulation of the Reaction Process of MIP Lifting Pipe Based on a Multi-Scale Model" proposed by Lubona, Cheng Congli, Lu Weimin, etc. in 2013. The parameters in the model were fitted according to the same test data as in the experimental example. The comparison between the fitted values and the test values is shown in Table 3 below.

[0229] Comparative Example 4

[0230] This comparative example uses an evaluation method and device for the yield of catalytic cracking reaction products (Patent No.: CN116072233A) to establish a product prediction model with a kinetic model. The parameters in the model were fitted according to the same test data as in the experimental example. The comparison between the fitted values and the test values is shown in Table 3 below.

[0231] From the comparison of the data in Table 3, it can be seen that the relative error of the product distribution in Experimental Example 1 is significantly smaller than that in Comparative Example 1, Comparative Example 2, Comparative Example 3, and Comparative Example 4, indicating that the modeling method proposed by the present invention has higher prediction accuracy; the relative error of the product distribution in Experimental Example 2 is smaller than that in Experimental Example 1, indicating that the modeling method introducing catalyst fixed carbon further improves the prediction accuracy. Table 1 Properties of the feedstock oil Table 2 Product distribution

[0232] Table 3 Average relative error of test values and fitted values Project Experimental Example 1 Experimental Example 2 Comparative Example 1 Comparative Example 2 Comparative Example 3 Comparative Example 4 Product Distribution, w% Dry Gas 1.52 1.05 3.97 4.49 5.24 3.65 Liquefied Gas 1.16 0.83 4.51 4.19 4.46 3.50 Gasoline 0.08 0.02 1.18 1.79 2.08 1.35 Diesel Oil 0.55 0.36 2.28 2.74 3.95 2.32 Slurry Oil 0.85 0.55 5.19 5.28 5.88 4.26 Coke 0.98 0.43 3.95 4.18 4.67 3.31

[0233] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for establishing a prediction model of a series reactor based on two reaction networks, wherein the series reactor includes a riser reactor and a fluidized bed reactor, and the raw material flows out after passing through the riser reactor and the fluidized bed reactor in sequence and reacting to generate a product; S1. According to the lumping kinetics principle and reaction mechanism, lump the raw material and the product and establish a reaction network; S2. Establish a kinetic model based on the reaction network; S3. Establish a prediction model according to the kinetic model; It is characterized in that: The reaction network includes a first reaction network and a second reaction network. The lumping of the first reaction network is as follows: According to the distillation range and the reaction mechanism of the riser reactor, the raw material in the riser reactor is divided into a light oil layer, a wax oil layer, and a residue oil layer, and is further divided into a light oil non-aromatic lumping, a light oil aromatic lumping, a wax oil non-aromatic lumping, a wax oil aromatic lumping, a residue oil non-aromatic, and a residue oil aromatic. The products of the riser reactor are divided into a dry gas lumping, a liquefied gas lumping, a gasoline saturated hydrocarbon lumping, a gasoline olefin lumping, a gasoline aromatic lumping, and a coke lumping; The lumping of the second reaction network is as follows: The raw material in the fluidized bed reactor is the product in the riser reactor. According to the distillation range and the reaction mechanism of the fluidized bed reactor, the raw material in the fluidized bed reactor is divided into a heavy oil layer, a light oil layer, and a gasoline layer, and is further divided into a heavy oil non-aromatic lumping, a heavy oil aromatic lumping, a light oil non-aromatic lumping, a light oil aromatic lumping, a gasoline saturated hydrocarbon lumping, a gasoline olefin lumping, and a gasoline aromatic lumping. The products in the fluidized bed reactor are divided into a dry gas lumping, a liquefied gas lumping, and a coke lumping.

2. The method for establishing a prediction model of a series reactor based on two reaction networks according to claim 1, characterized in that: The established first reaction network is as follows: The residue oil non-aromatic lumping reacts with the wax oil non-aromatic lumping, the light oil non-aromatic lumping, the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively; The wax oil non-aromatic lumping reacts with the light oil non-aromatic lumping, the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively; The gasoline non-aromatic lumping reacts with the dry gas lumping, the liquefied gas lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively; The residue oil aromatic lumping reacts with the wax oil aromatic lumping, the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively; The wax oil aromatic lumping reacts with the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively; The light oil aromatic lumping reacts with the dry gas lumping, the liquefied gas lumping, the gasoline aromatic lumping, and the coke lumping respectively.

3. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 1, characterized in that: The established second reaction network is as follows: The heavy oil non-aromatic lumping reacts with the light oil non-aromatic lumping, the gasoline saturated hydrocarbon lumping, and the gasoline olefin lumping respectively; The light oil non-aromatic lumping reacts with the gasoline saturated hydrocarbon lumping, the gasoline olefin lumping, the gasoline aromatic lumping, the dry gas lumping, and the liquefied gas lumping respectively; The heavy oil aromatic lumping reacts with the light oil aromatic lumping, the gasoline aromatic lumping, and the coke lumping respectively; The light oil aromatic lumping reacts with the gasoline aromatic lumping, the dry gas lumping, the liquefied gas lumping, and the coke lumping respectively; The gasoline saturated hydrocarbon lumping reacts with the dry gas lumping and the liquefied gas lumping respectively; The gasoline olefin lumping reacts with the gasoline saturated hydrocarbon lumping, the dry gas lumping, the liquefied gas lumping, and the gasoline aromatic lumping respectively; The gasoline aromatics lumps react with the dry gas lump, the liquefied gas lump and the coke lump respectively.

4. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 1, characterized in that: The prediction models in S3 include a riser reactor prediction model and a fluidized bed prediction model. The riser prediction model is: Among them, X Riser =x Riser / H Riser Bed layer x Riser Dimensionless relative distance at the cross section, H Riser represents the riser reactor height, y i is the mass fraction of each component, x Riser It represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the ith cluster in the riser reactor, Y Riser is the mass fraction vector of each lumped component in the riser reactor, S WH is the real weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, and P is the pressure in the riser reactor (Pa); The fluidized bed reactor prediction model is: Among them, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed cross-section in the fluidized bed layer, x bed is the distance from the fluidized bed inlet into the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the fixed carbon of the catalyst in the fluidized bed reactor, k W is the influence constant of the fixed carbon of the catalyst in the fluidized bed (cm 3 / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), S WH is the true weight hourly space velocity, y i is the mass fraction of each component, M i is the average relative molecular weight of the i-th lumping in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, Y Bed is the mass fraction vector of each lumping component in the fluidized bed reactor.

5. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 1, characterized in that: The kinetic models in S2 include a riser reactor kinetic model and a fluidized bed reactor kinetic model.

6. The method for establishing a prediction model of a series reactor based on two reaction networks according to claim 5, characterized in that: The riser kinetic model is derived from the riser continuity equation and the reaction rate: The riser continuity equation is: In the first reaction network, the reaction rate of the jth reaction is: The riser kinetic model is: Among them, ρ represents the density of the oil-gas mixture in the riser reactor (g / cm 3 ), t represents the reaction time, G v represents the mass flow rate of the oil-gas cross-section in the riser reactor (g / (cm 2 ·h)), x Riser represents the distance from the riser inlet into the reactor, a i is the concentration of the i-th lumping (molesi / g gas), R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the riser reactor, v i,j is the stoichiometric coefficient of the i-th lumping in the j-th reaction, r j represents the rate of the j-th reaction, P is the pressure in the riser reactor (Pa), k j is the reaction rate constant of the j-th reaction (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the riser reactor (g / cm 3 ), ε is the porosity, P is the system pressure of the riser reactor (Pa), R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the riser reactor (K), a = [a1,..., a 17 T is the lumped component concentration vector.​ 7. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 5, characterized in that: The fluidized bed kinetic model is derived from the fluidized bed reactor continuity equation and the reaction rate: The fluidized bed reactor continuity equation is: In the second reaction network, the reaction rate of the jth reaction is: The kinetic model of the fluidized bed reactor is: Among them, ρ represents the density of the oil-gas mixture in the fluidized bed reactor (g / cm 3 ), t represents the reaction time, G v represents the mass flow rate of the oil-gas cross-section in the fluidized bed reactor (g / (cm 2 ·h)), x Riser represents the distance from the fluidized bed inlet into the reactor, a i is the concentration of the i-th lumping (molesi / g gas), R i represents the reaction rate of the i-th lumping, n r is the number of reactions in the fluidized bed reactor, v i,j is the stoichiometric coefficient of the i-th lumping in the reaction j, r j represents the rate of the j-th reaction, P is the pressure in the fluidized bed reactor (Pa), k j is the reaction rate constant of the reaction j (cm 3 / (g·h)), ρ c is the catalyst density relative to the volume of the fluidized bed reactor (g / cm 3 ), ε is the porosity, P is the system pressure of the fluidized bed reactor (Pa), R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), a = [a1,..., a 17 T is the lumped component concentration vector.​ 8. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 1, characterized in that: The prediction models in S3 include a riser reactor prediction model and a fluidized bed reactor prediction model. Both the riser reactor prediction model and the fluidized bed reactor prediction model introduce a catalyst carbon content function to characterize its influence on the yield. The catalyst carbon content function in the riser reactor prediction model is: θ(C Riser ) = exp(-k W ·C Riser ) Among them, C Riser is the fixed carbon (w%) of the catalyst in the riser reactor, and k W is the influence constant of the fixed carbon of the catalyst; where y CK is the fraction of coke in the riser reactor, and R CO is the catalyst-oil ratio in the riser reactor; The catalyst carbon content function in the fluidized bed reactor prediction model is: θ(C Bed ) = exp(-k W ·C Bed ) Among them, C Bed is the fixed carbon (w%) of the catalyst in the fluidized bed reactor, and k W is the influence constant of the fixed carbon of the catalyst; Among them, R CO represents the catalyst-to-oil ratio in the fluidized bed reactor, y CK,Riser,out is the coke fraction at the outlet of the riser reactor, and β is the influence degree of reactor structure, reaction conditions, and feedstock oil properties on the fixed carbon of the catalyst.

9. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 8, characterized in that: The prediction model of the riser reactor is: Among them, X riser = x Riser / H Riser is the dimensionless relative distance at the x Riser cross-section in the bed layer, and H Riser represents the height of the riser reactor, y i is the mass fraction of each component, and x Riser represents the distance from the riser inlet into the reactor, M i is the average relative molecular weight of the i-th lumping in the riser reactor, Y Riser is the vector of the mass fractions of each lumping component in the riser reactor, S WH is the true weight hourly space velocity, K is the reaction rate constant matrix in the riser reactor, exp(-k W ·C Riser ) is the catalyst carbon deposition function in the riser reactor, and P is the pressure (Pa) in the riser reactor.

10. A method for establishing a prediction model of a series reactor based on two reaction networks according to claim 8, characterized in that: The prediction model of the fluidized bed reactor is: Among them, X Bed = x Bed / H Bed is the dimensionless relative distance at the x bed cross-section in the fluidized bed layer, x bed is the distance from the fluidized bed inlet into the fluidized bed reactor, H bed is the height of the fluidized bed reactor, C bed is the fixed carbon of the catalyst in the fluidized bed reactor, k W is the fixed carbon influence constant of the catalyst in the fluidized bed (cm 3 / (g / h)), P is the system pressure in the fluidized bed reactor, R is the gas constant (8.314 J / (mol·K)), T is the system temperature in the fluidized bed reactor (K), S WH is the true weight hourly space velocity, y i is the mass fraction of each component, M i is the average relative molecular weight of the i-th lumping in the fluidized bed reactor, K is the reaction rate constant matrix in the fluidized bed reactor, Y Bed is the mass fraction vector of each lumping component in the fluidized bed reactor.

11. An apparatus for establishing a prediction model of a series reactor based on two reaction networks, characterized in that: The device includes a modeling module that models using the riser and fluidized bed tandem catalytic cracking reaction model described in any one of claims 1-10.

12. An electronic device, characterized in that: It includes a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the prediction model described in any one of claims 1-10 can be run.

13. A readable storage medium, on which a computer program is stored, characterized in that: When the computer program is executed by the processor, the steps of the method described in any one of claims 1-10 are run.

Images