Esterification reactor based on solid acid catalyst

Abstract

The present invention relates to the technical field of esterification reactors, particularly to an esterification reactor based on a solid acid catalyst, the esterification reactor comprises a reactor tank body and a feed pipe, the reactor tank body is internally provided with an annular internal hollow structure catalytic filler bin, the catalytic filler bin is internally provided with a material turning assembly, and when a fixed shaft drives the catalytic filler bin to rotate, the material turning assembly turns over the material. The linkage assembly synchronously drives the material overturning assembly to work, so that the material overturning of the solid acid catalyst filled in the catalytic filler bin is synchronously completed when the catalytic filler bin rotates; according to the esterification reactor based on the solid acid catalyst, the contact uniformity of the solid acid catalyst and reactants is improved, partial excessive reaction of part of the solid acid catalyst is prevented, and the stability of esterification reaction is improved.

Description

Technical Field

[0001] The invention relates to the technical field of esterification reactors, in particular to an esterification reactor based on a solid acid catalyst. Background Art

[0002] The patent with publication number CN118416829A discloses an esterification reaction kettle, which includes a kettle body, a kettle cover, a material conveying assembly for conveying materials, a variable frequency stirring device for stirring materials, a temperature regulating device for regulating the temperature in the kettle body, and an esterification controller, wherein the material conveying assembly, the variable frequency stirring device, and the temperature regulating device are all connected to the esterification controller in communication; the upper opening of the kettle body is provided with a reaction chamber for material reaction, and the kettle cover is provided on the upper part of the kettle body; the variable frequency stirring device includes a variable frequency stirring motor, a mounting frame, a coupling, and a stirring paddle connected in sequence from top to bottom; the material conveying assembly includes a first feed pipe for conveying materials and a second feed assembly for conveying catalysts arranged on the kettle cover, and a discharge pipe arranged at the bottom of the kettle body for outputting reaction products. This application can improve the production efficiency of the esterification reaction and promote the stable progress of the esterification reaction.

[0003] Esterification is a common reaction in organic chemistry and is widely used in the fields of chemical industry, pharmaceuticals, fragrances, etc. For example, in the above application, traditional esterification reactions usually use liquid acids as catalysts, such as sulfuric acid, phosphoric acid, etc. However, liquid acid catalysts have certain corrosiveness and toxicity, and are difficult to separate after use, which can easily cause environmental pollution. Therefore, research and development of an environmentally friendly and efficient esterification catalyst has become an important topic.

[0004] As a new type of catalyst, solid acid catalyst has gradually become a research hotspot due to its good catalytic performance and easy separation. However, there are some problems with the existing esterification reactor based on solid acid catalyst, that is, the existing esterification reactor using solid acid catalyst fills the solid acid catalyst in the catalytic filling bin set in the reactor for use. Due to the uneven contact between the reaction raw materials and the solid acid catalyst in the catalytic filling bin, some solid acid catalysts are locally over-reacted, affecting the activity of the catalyst, thereby affecting the stability of the esterification reaction. Summary of the invention

[0005] In order to solve the above problems, the present invention provides an esterification reactor based on a solid acid catalyst.

[0006] The present invention adopts the following technical scheme: an esterification reactor based on a solid acid catalyst comprises a reactor tank body and a feed pipe, wherein a catalytic filling bin with an annular internal hollow structure is arranged in the reactor tank body, and the catalytic filling bin is filled with a solid acid catalyst, a fixed shaft is rotatably connected at the top center of the reactor tank body, and the catalytic filling bin is fixedly connected to the fixed shaft, a flow equalizing plate is arranged at the upper part of the catalytic filling bin in the reactor tank body, a linkage assembly is arranged below the flow equalizing plate, and a material turning assembly is arranged in the catalytic filling bin, when the fixed shaft drives the catalytic filling bin to rotate, the linkage assembly synchronously drives the material turning assembly to work, so that the solid acid catalyst filled inside is turned synchronously when the catalytic filling bin rotates;

[0007] It also includes a reaction control module, which is arranged on the outer wall of the reactor tank, and the reaction control module includes:

[0008] A data acquisition module collects historical control training parameters of the esterification reactor. The historical control training parameters are collected when the raw materials are in normal reaction. The historical control training parameters include comprehensive reaction parameters and feed flow rate. The comprehensive reaction parameters include reaction coefficient, catalyst conversion rate and reactant type.

[0009] Wherein, the parameters affecting the reaction coefficient include raw material heating temperature, reaction pressure and reactant concentration;

[0010] The expression of the reaction coefficient is:

[0011] ;

[0012] In the formula, is the reaction coefficient, is the raw material heating temperature, is the reaction pressure, is the type of reactant, , and is the weight factor, , and Both are greater than zero.

[0013] Model training module, based on historical control training parameters, trains a machine learning model to predict feed flow, collects real-time comprehensive reaction parameters, and predicts feed flow based on the trained machine learning model;

[0014] The controller generates a flow adjustment instruction based on the predicted feed flow, and adjusts the flow control valve on the feed pipe based on the flow adjustment instruction.

[0015] As a further description of the above technical solution: the material turning assembly includes a driving rod and a spiral blade, the driving rod is rotatably connected to the inner top plate of the catalytic filling bin, and the spiral blade is integrally formed on the outer wall of the driving rod;

[0016] The linkage assembly includes an annular gear ring, a connecting block and a driving shaft. The annular gear ring is fixedly connected to the flow equalizing plate through the connecting block. The driving shaft is rotatably connected to the upper surface of the catalytic filling bin and is fixedly connected to the driving rod. A linkage gear is welded to the top end of the driving shaft, and the linkage gear is meshed with the annular gear ring gear.

[0017] As a further description of the above technical solution: the catalytic filling bin and the flow equalizing plate are both hollow structures.

[0018] As a further description of the above technical solution: a bracket is welded to the bottom of the outer wall of the reactor tank body, a mounting seat is welded on the bracket, a heat exchanger body is mounted on the mounting seat, a heat exchange coil is arranged in the heat exchanger body, a feed pipe is arranged on the reactor tank body, the feed pipe is conductively connected to one end of the heat exchange coil, and the other end of the heat exchange coil passes through the heat exchanger body and extends to the outside, and a discharge pipe is also arranged on the reactor tank body, and the discharge pipe is conductively connected to the heat exchanger body.

[0019] As a further description of the above technical solution: a motor seat is welded at the top center of the reactor tank, a driving motor is bolted to the motor seat, and an output shaft of the driving motor is fixedly connected to the fixed shaft through a coupling.

[0020] As a further description of the above technical solution: the training method of the machine learning model for training and predicting the feed flow rate includes:

[0021] Convert the collected historical control training parameters into a corresponding set of feature vectors;

[0022] Each group of feature vectors is used as the input of the machine learning model, and the machine learning model takes the feed flow rate corresponding to each group of reaction comprehensive parameters as the output, and the feed flow rate actually corresponding to each group of reaction comprehensive parameters as the prediction target, and minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0023] As a further description of the above technical solution: the method for obtaining the catalyst conversion rate includes:

[0024] Pre-collecting historical activity training data required for predicting catalyst conversion rate, wherein the historical activity training data is collected in an experimental environment, wherein the experimental environment is an environment in which testers continuously collect real-time conversion rate data after using the catalyst reaction by controlling physical factors that affect the catalyst conversion rate;

[0025] The historical activity training data includes i groups of activity characteristic parameters and catalyst conversion rate, wherein the activity characteristic parameters include real-time catalyst temperature, catalyst use time, reactant usage, and catalyst size and type;

[0026] A second machine learning model for predicting catalyst conversion rate is trained based on historical activity training data, and the activity characteristic parameters of the catalyst used are collected in real time. The catalyst conversion rate is predicted based on the trained machine learning model for predicting catalyst conversion rate.

[0027] As a further description of the above technical solution: the method for training a second machine learning model for predicting catalyst conversion rate includes:

[0028] Converting the collected historical activity training data into a corresponding set of second feature vectors;

[0029] Each group of second feature vectors is used as the input of the second machine learning model. The second machine learning model takes the catalyst conversion rate corresponding to each group of activity characteristic parameters as the output, takes the catalyst conversion rate actually corresponding to each group of activity characteristic parameters as the prediction target, and takes minimizing the loss function value of the second machine learning model as the training target; training is stopped when the loss function value of the second machine learning model is less than or equal to the preset second target loss value.

[0030] Beneficial effects:

[0031] The esterification reactor based on solid acid catalyst provided by the present invention, when the esterification reactor based on solid acid catalyst is used, firstly, when the reactant is fed into the reactor tank body through the feed pipe, the reactant fed into the reactor tank body through the feed pipe can be dispersed by setting the flow equalizing plate, so that it can fall evenly through the hollow flow equalizing plate without flowing in strands, which can disperse the reactant and improve the uniformity of its mixing, and secondly improve the uniformity of the contact between the reactant and the solid acid catalyst filled in the catalytic filling bin, and secondly, when the fixed shaft is controlled to rotate by the driving motor, the fixed shaft is synchronous The catalytic filling bin is driven to rotate, thereby improving the uniformity of contact between the reactants and the solid acid catalyst filled in the catalytic filling bin. In conjunction with the setting of the linkage component, when the catalytic filling bin is driven by the fixed shaft to rotate, the linkage component synchronously drives the turning component to work, so that when the catalytic filling bin rotates, the solid acid catalyst filled inside is turned over synchronously, that is, the solid acid catalyst located at the bottom of the catalytic filling bin is transported to the top for turning and mixing, thereby further improving the uniformity of contact between the solid acid catalyst and the reactants, preventing the occurrence of local overreaction of some solid acid catalysts, and improving the stability of the esterification reaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0033] Figure 1 A schematic diagram of the structure of an esterification reactor based on a solid acid catalyst provided in an embodiment of the present invention;

[0034] Figure 2 A cross-sectional view of an esterification reactor based on a solid acid catalyst provided in an embodiment of the present invention;

[0035] Figure 3 A schematic diagram of the structure of a flow balancing plate and a catalytic filling bin provided in an embodiment of the present invention;

[0036] Figure 4 A schematic diagram of the structure of a current equalizing plate provided in an embodiment of the present invention;

[0037] Figure 5 A cross-sectional view of a catalytic filling bin provided in an embodiment of the present invention.

[0038] Figure numerals: 1. Reactor tank; 101. Motor base; 102. Driving motor; 103. Fixed shaft; 104. Flow equalizing plate; 105. Catalytic filling bin; 106. Linkage assembly; 1061. Connecting block; 1062. Annular gear ring; 1063. Driving shaft; 1064. Linkage gear; 107. Turning assembly; 1071. Driving rod; 1072. Spiral blade; 2. Bracket; 3. Discharge pipe; 5. Feed pipe; 6. Heat exchanger body; 61. Heat exchange coil; 7. Mounting base. DETAILED DESCRIPTION

[0039] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention is further described below with reference to specific diagrams. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0040] Example 1

[0041] See also Figure 1-Figure 5The embodiment of the present invention provides a technical solution: an esterification reactor based on a solid acid catalyst, comprising a reactor tank body 1 and a feed pipe 5, wherein a catalytic filling bin 105 with an annular internal hollow structure is arranged in the reactor tank body 1, and the catalytic filling bin 105 is filled with a solid acid catalyst, and a fixed shaft 103 is rotatably connected at the top center of the reactor tank body 1, and the catalytic filling bin 105 is fixedly connected to the fixed shaft 103. Specifically, when in use, the catalytic filling bin 105 can be driven to rotate by the fixed shaft 103 to improve the uniformity of contact between the reactant and the solid acid catalyst filled in the catalytic filling bin 105; a flow balancing plate 104 is arranged on the upper part of the catalytic filling bin 105 in the reactor tank body 1, and specifically, the setting of the flow balancing plate 104 can disperse the reactant fed into the reactor tank body 1 through the feed pipe 5, so that it falls evenly through the hollow flow balancing plate 104 without flowing in strands, which can improve the uniformity of the reaction. The materials are dispersed to improve the uniformity of mixing, and secondly, the uniformity of contact between the reactants and the solid acid catalyst filled in the catalytic filling bin 105 is improved; a linkage assembly 106 is arranged below the flow equalizing plate 104, and a flipping assembly 107 is arranged in the catalytic filling bin 105. When the fixed shaft 103 drives the catalytic filling bin 105 to rotate, the flipping assembly 107 is synchronously driven by the linkage assembly 106 to work, so that the flipping of the solid acid catalyst filled inside is completed synchronously when the catalytic filling bin 105 rotates; the flipping assembly 107 includes a driving rod 1071 and a spiral blade 1072, the driving rod 1071 is rotatably connected to the internal top plate of the catalytic filling bin 105, and the spiral blade 1072 is integrally formed on the outer wall of the driving rod 1071; specifically, there are three flipping assemblies 107, and the three flipping assemblies 107 are indirectly distributed in a ring shape, and the spacing between two adjacent flipping assemblies 107 is equal. The linkage assembly 106 includes an annular gear ring 1062, a connecting block 1061 and a drive shaft 1063. The annular gear ring 1062 is fixedly connected to the flow equalizing plate 104 through the connecting block 1061. The drive shaft 1063 is rotatably connected to the upper surface of the catalytic filling bin 105 and is fixedly connected to the drive rod 1071. A linkage gear 1064 is welded to the top of the drive shaft 1063. The linkage gear 1064 is meshed with the annular gear ring 1062. The catalytic filling bin 105 and the flow equalizing plate 104 are both hollow structures. A motor seat 101 is welded at the top center of the reactor tank 1. A drive motor 102 is bolted to the motor seat 101. The output shaft of the drive motor 102 is fixedly connected to the fixed shaft 103 through a coupling. Specifically, the bottom end of the fixed shaft 103 is located below the catalytic filling bin 105 and is provided with a stirring blade. The stirring blade is not shown in the figure. When the fixed shaft 103 is driven to rotate by the driving motor 102, the stirring blade can be driven to rotate to stir and mix the raw materials.

[0042] In this embodiment, when the esterification reactor based on the solid acid catalyst is in use, first, when the reactants are fed into the reactor tank body 1 through the feed pipe 5, the flow equalizing plate 104 is set to disperse the reactants fed into the reactor tank body 1 through the feed pipe 5, so that they fall evenly through the hollow flow equalizing plate 104 without flowing in strands. On the one hand, the reactants can be dispersed to improve the uniformity of mixing, and secondly, the uniformity of contact between the reactants and the solid acid catalyst filled in the catalytic filling bin 105 is improved. Secondly, when the fixed shaft 103 is controlled to rotate by the driving motor 102, the fixed shaft 103 synchronously drives the catalytic filling bin 105 to rotate. The catalytic filling bin 105 is movable to improve the uniformity of contact between the reactants and the solid acid catalyst filled in the catalytic filling bin 105, and in conjunction with the setting of the linkage component 106, when the catalytic filling bin 105 is driven by the fixed shaft 103 to rotate, the linkage component 106 synchronously drives the turning component 107 to work, so that when the catalytic filling bin 105 rotates, the solid acid catalyst filled inside is turned over synchronously, that is, the solid acid catalyst located at the bottom of the catalytic filling bin 105 is transported to the top for turning and mixing, thereby further improving the uniformity of contact between the solid acid catalyst and the reactants, preventing the occurrence of local overreaction of some solid acid catalysts, and improving the stability of the esterification reaction.

[0043] Example 2

[0044] Reference Figure 1-Figure 2 Based on the above embodiment, this embodiment adds a heat exchanger body 6 and a heat exchange coil 61 to preheat the reactants, thereby improving the reaction efficiency and realizing the recovery and utilization of the heat energy generated by the esterification reaction;

[0045] A bracket 2 is welded to the bottom of the outer wall of the reactor tank body 1, and a mounting seat 7 is welded on the bracket 2. A heat exchanger body 6 is installed on the mounting seat 7. A heat exchange coil 61 is arranged in the heat exchanger body 6. A feed pipe 5 is arranged on the reactor tank body 1. The feed pipe 5 is conductively connected to one end of the heat exchange coil 61, and the other end of the heat exchange coil 61 passes through the heat exchanger body 6 and extends to the outside. A discharge pipe 3 is also arranged on the reactor tank body 1, and the discharge pipe 3 is conductively connected to the heat exchanger body 6.

[0046] In this embodiment, the products of the esterification reaction (gaseous water and fat) are discharged into the heat exchanger body 6 through the discharge pipe 3, and the reactants are discharged into the reactor tank body 1 through the feed pipe 5 after passing through the heat exchange coil 61, so that the reactants exchange heat with the products of the esterification reaction through the heat exchange coil 61, and the reactants are preheated to increase the temperature of the reactants when entering the reactor tank body 1, thereby accelerating the reaction speed. The heat exchange coil 61 is spiral, so that the reactants can fully absorb heat during the preheating process, and at the same time increase the flow path to ensure uniform temperature.

[0047] Example 3

[0048] The solid acid catalyst-based esterification reactor further includes a reaction control module, which is disposed on the outer wall of the reactor tank 1. The reaction control module includes:

[0049] A data acquisition module collects historical control training parameters of the esterification reactor. The historical control training parameters are collected when the raw materials are in normal reaction. The historical control training parameters include comprehensive reaction parameters and feed flow rate. The comprehensive reaction parameters include reaction coefficient, catalyst conversion rate and reactant type.

[0050] Model training module, based on historical control training parameters, trains a machine learning model to predict feed flow, collects real-time comprehensive reaction parameters, and predicts feed flow based on the trained machine learning model;

[0051] The controller generates a flow adjustment instruction based on the predicted feed flow, and adjusts the flow control valve on the feed pipe 5 based on the flow adjustment instruction.

[0052] Parameters that affect the reaction coefficient include raw material heating temperature, reaction pressure and reactant concentration;

[0053] Specifically, the types of reactants are directly obtained through raw material labels, and the raw material heating temperature and reaction pressure are collected and obtained through a temperature sensor and a pressure sensor disposed in the reactor tank 1 .

[0054] The expression of the reaction coefficient is:

[0055] ;

[0056] In the formula, is the reaction coefficient, is the raw material heating temperature, is the reaction pressure, is the type of reactant, , and is the weight factor, , and Both are greater than zero.

[0057] It should be noted that the size of the weight factor is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight factor depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight factor for each set of comprehensive parameters by technical personnel in this field.

[0058] It should be noted that temperature is an important factor affecting the esterification reaction rate. Within a suitable range, increasing the temperature can increase the reaction rate and yield, that is, the higher the temperature, the faster the reaction efficiency, so the feed flow rate can be increased, and vice versa; in the esterification reaction, the use of pressurization can increase the boiling point of the reactants, promote the liquid phase reaction, and increase the product yield, that is, the greater the reaction pressure, the faster the reaction efficiency, so the feed flow rate can be increased, and vice versa; the feed rate is usually inversely proportional to the concentration of the reactants, that is, the higher the concentration of the reactants, the slower the feed rate, and vice versa. In summary, the greater the reaction coefficient, the greater the feed flow rate, and vice versa.

[0059] The training method for training a machine learning model to predict feed flow rate includes:

[0060] Convert the collected historical control training parameters into a corresponding set of feature vectors;

[0061] Each set of feature vectors is used as the input of the machine learning model. The machine learning model takes the feed flow rate corresponding to each set of reaction comprehensive parameters as the output, the feed flow rate actually corresponding to each set of reaction comprehensive parameters as the prediction target, and minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0062] The machine learning model may be one of support vector machine regression, random forest regression or neural network regression models.

[0063] The loss function value of the machine learning model is the mean square error.

[0064] Mean square error is one of the commonly used loss functions. The model is trained with minimization as the goal, so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model.

[0065] In the loss function is the loss function value of the machine learning model, is the feature vector group number; is the number of eigenvector groups; For the The feed flow rate corresponding to the group eigenvector, For the The feed flow rate actually corresponding to the group characteristic vector.

[0066] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0067] In this embodiment, the feed flow rate is predicted based on the trained machine learning model according to the real-time comprehensive reaction parameters collected, and then a flow adjustment instruction is generated based on the predicted feed flow rate. The flow control valve on the feed pipe 5 is adjusted based on the flow adjustment instruction, thereby realizing automatic real-time dynamic adjustment of the feed flow rate, thereby ensuring that the reaction is carried out effectively and stably, and improving energy utilization efficiency.

[0068] Example 4

[0069] Based on the above embodiments, this embodiment further discloses a method for obtaining the catalyst conversion rate. It should be noted that the catalyst conversion rate, i.e., the ratio of reactants converted to products, is calculated based on the changes in the initial concentration of the reactants and the concentration after the reaction.

[0070] The method for obtaining the catalyst conversion rate includes:

[0071] Collect in advance the historical activity training data required for predicting the catalyst conversion rate. The historical activity training data is collected in an experimental environment. The experimental environment is an environment in which testers continuously collect real-time conversion rate data after using the catalyst reaction by controlling the physical factors that affect the catalyst conversion rate.

[0072] The historical activity training data includes i groups of activity characteristic parameters and catalyst conversion rate, and the activity characteristic parameters include the real-time temperature of the catalyst, the catalyst use time, the amount of reactants used, and the size and type of the catalyst;

[0073] A second machine learning model for predicting catalyst conversion rate is trained based on historical activity training data, and activity characteristic parameters of the catalyst used are collected in real time. The catalyst conversion rate is predicted based on the trained second machine learning model for predicting catalyst conversion rate.

[0074] The method of training a second machine learning model for predicting catalyst conversion includes:

[0075] Converting the collected historical activity training data into a corresponding set of second feature vectors;

[0076] Each group of second eigenvectors is used as the input of the second machine learning model. The second machine learning model takes the catalyst conversion rate corresponding to each group of activity characteristic parameters as the output, takes the catalyst conversion rate actually corresponding to each group of activity characteristic parameters as the prediction target, and takes minimizing the loss function value of the second machine learning model as the training target; training is stopped when the loss function value of the second machine learning model is less than or equal to the preset target loss value.

[0077] The second machine learning model can be one of support vector machine regression, random forest regression, neural network regression and other models. The loss function value of the second machine learning model is mean square error, which is one of the commonly used loss functions. By Minimization is used as the goal to train the model so that the second machine learning model better fits the data, thereby improving the performance and accuracy of the model;

[0078] In the loss function is the loss function value of the second machine learning model, is the second eigenvector group number; is the number of second eigenvector groups; For the The second eigenvector of the group predicts the catalyst conversion rate, For the The catalyst conversion rate actually corresponding to the second eigenvector of the group; the other model parameters of the second machine learning model, the target loss value, the optimization algorithm, the ratio of the training set, the test set, the validation set, and the optimization of the loss function are all achieved through actual engineering and obtained after continuous experimental tuning.

[0079] In this embodiment, historical activity training data required for predicting catalyst conversion rate is collected in advance, and the historical activity training data includes i groups of activity characteristic parameters and catalyst conversion rate, and the activity characteristic parameters include real-time catalyst temperature, catalyst usage time, reactant usage, and catalyst size and type. A second machine learning model for predicting catalyst conversion rate is trained based on the historical activity training data, and then the activity characteristic parameters of the used catalyst are collected in real time. The catalyst conversion rate is predicted based on the trained second machine learning model for predicting catalyst conversion rate, so that the catalyst conversion rate, that is, the activity of the catalyst, can be obtained in real time. The feed flow rate is dynamically adjusted according to the catalyst conversion rate to ensure the stable progress of the esterification reaction.

[0080] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and the description in the specification are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An esterification reactor based on a solid acid catalyst, comprising a reactor tank (1) and a feed pipe (5), characterized in that: The reactor tank (1) is provided with a catalytic filling bin (105) having an annular internal hollow structure, which is filled with a solid acid catalyst. A fixed shaft (103) is rotatably connected at the top center of the reactor tank (1), and the catalytic filling bin (105) is fixedly connected to the fixed shaft (103). A flow equalizing plate (104) is provided at the upper part of the catalytic filling bin (105) in the reactor tank (1), and a linkage assembly (106) is provided below the flow equalizing plate (104). A material turning assembly (107) is provided in the catalytic filling bin (105). When the fixed shaft (103) drives the catalytic filling bin (105) to rotate, the linkage assembly (106) synchronously drives the material turning assembly (107) to work, so that the solid acid catalyst filled in the catalytic filling bin (105) is turned synchronously when the catalytic filling bin (105) rotates. It also includes a reaction control module, which is arranged on the outer wall of the reactor tank (1), and the reaction control module includes: A data acquisition module collects historical control training parameters of the esterification reactor. The historical control training parameters are collected when the raw materials are in normal reaction. The historical control training parameters include comprehensive reaction parameters and feed flow rate. The comprehensive reaction parameters include reaction coefficient, catalyst conversion rate and reactant type. Among them, the parameters that affect the reaction coefficient include raw material heating temperature, reaction pressure and reactant concentration; The expression of the reaction coefficient is: ； In the formula, is the reaction coefficient, is the raw material heating temperature, is the reaction pressure, is the type of reactant, , and is the weight factor, , and are greater than zero; Model training module, based on historical control training parameters, trains a machine learning model to predict feed flow, collects real-time comprehensive reaction parameters, and predicts feed flow based on the trained machine learning model; The controller generates a flow adjustment instruction based on the predicted feed flow rate, and adjusts the flow control valve on the feed pipe (5) based on the flow adjustment instruction.

2. The esterification reactor based on a solid acid catalyst according to claim 1, characterized in that: The material turning assembly (107) comprises a driving rod (1071) and a spiral blade (1072); the driving rod (1071) is rotatably connected to the inner top plate of the catalytic filling bin (105); and the spiral blade (1072) is integrally formed on the outer wall of the driving rod (1071); The linkage assembly (106) comprises an annular gear ring (1062), a connecting block (1061) and a driving shaft (1063); the annular gear ring (1062) is fixedly connected to the flow equalizing plate (104) via the connecting block (1061); the driving shaft (1063) is rotatably connected to the upper surface of the catalytic filling bin (105) and is fixedly connected to the driving rod (1071); a linkage gear (1064) is welded to the top end of the driving shaft (1063); and the linkage gear (1064) is meshingly connected to the annular gear ring (1062).

3. The esterification reactor based on a solid acid catalyst according to claim 1, characterized in that: The catalytic filling bin (105) and the flow equalizing plate (104) are both hollow structures.

4. The esterification reactor based on a solid acid catalyst according to claim 1, characterized in that: A bracket (2) is welded to the bottom of the outer wall of the reactor tank body (1), a mounting seat (7) is welded to the bracket (2), a heat exchanger body (6) is mounted on the mounting seat (7), a heat exchange coil (61) is arranged in the heat exchanger body (6), the feed pipe (5) is conductively connected to one end of the heat exchange coil (61), and the other end of the heat exchange coil (61) passes through the heat exchanger body (6) and extends to the outside, and a discharge pipe (3) is also arranged on the reactor tank body (1), and the discharge pipe (3) is conductively connected to the heat exchanger body (6).

5. The esterification reactor based on a solid acid catalyst according to claim 1, characterized in that: A motor seat (101) is welded at the center of the top of the reactor tank (1), a driving motor (102) is bolted to the motor seat (101), and an output shaft of the driving motor (102) is fixedly connected to a fixed shaft (103) via a coupling.

6. The esterification reactor based on solid acid catalyst according to claim 1, characterized in that: The training method of the machine learning model for training and predicting feed flow rate includes: Convert the collected historical control training parameters into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the machine learning model, and the machine learning model takes the feed flow rate corresponding to each group of reaction comprehensive parameters as the output, and the feed flow rate actually corresponding to each group of reaction comprehensive parameters as the prediction target, and minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

7. The esterification reactor based on a solid acid catalyst according to claim 6, characterized in that: The method for obtaining the catalyst conversion rate comprises: Collect in advance the historical activity training data required to predict the catalyst conversion rate; The historical activity training data includes i groups of activity characteristic parameters and catalyst conversion rate, wherein the activity characteristic parameters include real-time catalyst temperature, catalyst use time, reactant usage, and catalyst size and type; A second machine learning model for predicting catalyst conversion rate is trained based on historical activity training data, and activity characteristic parameters of the catalyst used are collected in real time. The catalyst conversion rate is predicted based on the trained second machine learning model for predicting catalyst conversion rate.

8. The esterification reactor based on a solid acid catalyst according to claim 7, characterized in that: The method of training a second machine learning model for predicting catalyst conversion rate comprises: Converting the collected historical activity training data into a corresponding set of second feature vectors; Each group of second feature vectors is used as the input of the second machine learning model. The second machine learning model takes the catalyst conversion rate corresponding to each group of activity characteristic parameters as the output, takes the catalyst conversion rate actually corresponding to each group of activity characteristic parameters as the prediction target, and takes minimizing the loss function value of the second machine learning model as the training target; training is stopped when the loss function value of the second machine learning model is less than or equal to the preset second target loss value.

Images