Single-screw extrusion reactor design method based on Bayesian optimization and application thereof

Abstract

The invention discloses a Bayesian optimization-based single-screw extrusion reactor design method and application thereof. The method comprises the following steps of: determining key structure parameters and process parameters of a single-screw extrusion reactor so as to establish an initial parameterized geometric model; importing the initial parameterized geometric model into simulation software for analog simulation to obtain performance parameters of the reactor model; a parameter communication interface of simulation software and a Bayesian optimization algorithm is realized through a script, performance parameters are used as a target function, and the Bayesian optimization algorithm is used for automatically optimizing the screw pitch, the screw rotating speed and the catalyst dosage of the screw; automatically updating parameters of the reactor model according to an optimization result, and performing simulation calculation again; iterative optimization is carried out until convergence, and finally, the structure parameters and the process parameters of the single-screw extrusion reactor with the optimal performance are obtained. The design method of the single-screw extrusion reactor is high in automation degree, good in optimization precision, high in optimization efficiency and good in application prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of reactor design, in particular to a single-screw extrusion reactor design method based on Bayesian optimization and its application, and in particular to a single-screw extrusion reactor structure and process parameter optimization design method based on Bayesian optimization and its application. Background Art

[0002] In the polymer processing industry, single-screw extrusion reactors are widely used in polymer modification and catalytic cracking reactions. With market demands for higher product quality and production efficiency, optimizing reactor structure and operating parameters has become a key issue. Existing single-screw extrusion reactor design methods typically rely on experience or limited experimentation, making it difficult to systematically optimize structural and process parameters. Furthermore, traditional optimization methods often require extensive repetitive experiments, which is time-consuming, labor-intensive, and inefficient.

[0003] In recent years, computational fluid dynamics (CFD) simulation technology has been widely used in reactor design, accurately simulating fluid flow and heat transfer during reactions. However, simple CFD simulation cannot automatically search for and determine the optimal parameter combination, requiring repeated manual intervention and trial and error.

[0004] Therefore, how to achieve automated and systematic optimization of the structure and process parameters of single-screw extrusion reactors has become an important issue that needs to be urgently addressed in this field. Summary of the Invention

[0005] Due to the above-mentioned defects in the prior art, the present invention provides an automatic and systematic method for optimizing the structure and process parameters of a single-screw extrusion reactor, specifically a method for optimizing the structure and process parameters of a single-screw extrusion reactor based on Bayesian optimization and its application, which overcomes the problems of the existing computational fluid dynamics simulation method of the single-screw extrusion reactor, which still requires repeated manual intervention and trial and error, is inefficient and lacks automation.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The single-screw extrusion reactor design method based on Bayesian optimization includes the following steps:

[0008] (1) Determine the key structural parameters and process parameters of the single-screw extrusion reactor, and establish an initial parametric geometric model of the single-screw extrusion reactor based on these parameters. The key structural parameters and process parameters include screw pitch, screw speed, and catalyst dosage.

[0009] (2) Importing the initial parameterized geometric model established in step (1) into the simulation software to simulate the LDPE catalytic cracking process and obtain the performance parameters of the reactor model;

[0010] (3) Implementing a parameter communication interface between the simulation software and the Bayesian optimization algorithm through a script, using the performance parameters obtained in step (2) as the objective function, and automatically optimizing the screw pitch, screw speed, and catalyst dosage using the Bayesian optimization algorithm, with the optimization direction being the optimal performance parameters;

[0011] (4) Automatically update the parameters of the reactor model according to the optimization results of step (3) and perform simulation calculations again;

[0012] (5) Repeat steps (3) and (4) until convergence, and finally obtain the structural parameters and process parameters of the single-screw extrusion reactor with optimal performance.

[0013] The single-screw extrusion reactor design method based on Bayesian optimization of the present invention has a reasonable step design. By combining the Bayesian optimization algorithm with simulation, intelligent automatic adjustment of parameters can be achieved, thereby realizing the automated optimization design of the structure and process parameters of the single-screw extrusion reactor, which can greatly improve the optimization efficiency and optimization accuracy and greatly reduce the simulation cost. The method can be widely applied to the optimization design of various polymer extrusion reactors, has wide applicability and good application prospects.

[0014] As the preferred technical solution:

[0015] In the single-screw extrusion reactor design method based on Bayesian optimization described above, the performance parameters include the mass flow rate of the reactant LDPE at the reactor outlet, the mass flow rate of the product gasoline, and the mass flow rate of the by-product coke. The performance parameter selected in step (3) can be any one of the above parameters, or all of the above parameters can be considered as needed. Generally, the minimum mass flow rate of the reactant LDPE at the reactor outlet can be used as the control target.

[0016] The single-screw extrusion reactor design method based on Bayesian optimization as described above, wherein the simulation software is CFD simulation software;

[0017] The script is a Python script.

[0018] As described above, in the single-screw extrusion reactor design method based on Bayesian optimization, the Python script code realizes the automated execution of the simulation process through the CFD parameter interface.

[0019] In the Bayesian optimization-based single-screw extrusion reactor design method described above, the simulation sets the catalytic cracking reaction as a steady-state continuous reaction process. Specifically, the CFD simulation uses an Euler two-phase flow model, with phase one being a liquid phase and phase two being a solid phase, and the flow being transient, continuous, and incompressible.

[0020] In the single-screw extrusion reactor design method based on Bayesian optimization as described above, the initial parameterized geometric model and CFD simulation boundary conditions of the single-screw extrusion reactor are both parameterized designs.

[0021] As described above, in the single-screw extrusion reactor design method based on Bayesian optimization, the automatic optimization process can complete the continuous iterative simulation of the reactor model without human intervention.

[0022] In the single-screw extrusion reactor design method based on Bayesian optimization, the catalyst dosage is the catalyst inlet volume fraction. The screw pitch ranges from 30 to 60 mm, the screw speed ranges from 80 to 140 rpm, and the catalyst inlet volume fraction ranges from 0.25 to 0.45.

[0023] The present invention further provides a computer device, comprising:

[0024] at least one processor; and,

[0025] a memory communicatively connected to the at least one processor; wherein,

[0026] The memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the single-screw extrusion reactor design method based on Bayesian optimization as described above is implemented.

[0027] In addition, the present invention also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the single-screw extrusion reactor design method based on Bayesian optimization as described above is implemented.

[0028] The above technical solution is only a feasible technical solution of the present invention. The protection scope of the present invention is not limited thereto. Those skilled in the art can reasonably adjust the specific design according to actual needs.

[0029] The above invention has the following advantages or beneficial effects:

[0030] (1) The single-screw extrusion reactor design method based on Bayesian optimization of the present invention has a reasonable step design. By combining the Bayesian optimization algorithm with CFD simulation, the intelligent automatic adjustment of parameters can be realized, thereby realizing the automated optimization design of the structure and process parameters of the single-screw extrusion reactor, which can greatly improve the optimization efficiency and optimization accuracy and greatly reduce the simulation cost.

[0031] (2) The single-screw extrusion reactor design method based on Bayesian optimization of the present invention can be widely applied to the optimization design of various polymer extrusion reactors, has wide applicability and good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The present invention and its features, configurations, and advantages will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings. Like reference numerals indicate like parts throughout the drawings. The drawings are not necessarily drawn to scale, emphasis being placed on illustrating the subject matter of the present invention.

[0033] Figure 1 A step diagram of a method for designing a single-screw extrusion reactor according to the present invention;

[0034] Figure 2 Schematic diagram of the structure of a single-screw extrusion reactor;

[0035] Figure 3 Optimize the result graph for the Bayesian optimization algorithm;

[0036] Figure 4 The line graph of LDPE mass fraction at the outlet under five different parameter conditions;

[0037] Figure 5 This is the LDPE mass fraction cloud diagram under the conditions of a screw pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1;

[0038] Figure 6 This is the cloud diagram of the mass fraction of the intermediate product LDPEC under the conditions of a pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1;

[0039] Figure 7 This is a cloud diagram of the mass fraction of product gasoline under the conditions of a pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1;

[0040] Figure 8 This is a cloud diagram of the mass fraction of light olefins in the product under the conditions of a pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1;

[0041] Figure 9 This is a cloud diagram of the product coke mass fraction under the conditions of a pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1;

[0042] Figure 10 This is the temperature cloud diagram inside the reactor under the conditions of a pitch of 60 mm, a rotation speed of 120 rpm, and a catalyst to raw material feed ratio of 1:1. DETAILED DESCRIPTION

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0044] Example 1

[0045] A single screw extrusion reactor design method based on Bayesian optimization, the steps are as follows Figure 1 As shown, the details are as follows:

[0046] (1) Determine Figure 2 The key structural parameters and process parameters of the single-screw extrusion reactor shown in the figure are used to establish the initial parameterized geometric model of the single-screw extrusion reactor. The key structural parameters and process parameters include screw pitch (30 mm), screw speed (120 rpm) and catalyst dosage (catalyst inlet volume fraction, specifically 0.25);

[0047] (2) The initial parameterized geometric model established in step (1) was imported into the CFD simulation software to perform CFD simulation of the LDPE catalytic cracking process. The conditions of the CFD simulation (the catalytic cracking reaction was set as a steady-state continuous reaction process) were a screw temperature of 500°C and an inlet material flow rate of 0.04 m / s, and the performance parameters of the reactor model were obtained (the performance parameters included the mass flow rate at the LDPE outlet, the mass flow rate at the product gasoline outlet, and the mass flow rate at the product coke outlet);

[0048] (3) The parameter communication interface between CFD and Bayesian optimization algorithm is realized through Python script, and the minimum mass flow rate at the LDPE outlet is used as the control target (of course, the maximum mass flow rate at the product gasoline outlet and the minimum mass flow rate at the product coke outlet can also be used as the control target). The Bayesian optimization algorithm is used to automatically optimize the screw pitch, screw speed and catalyst dosage;

[0049] (4) Automatically update the parameters of the reactor model according to the optimization results of step (3) and perform simulation calculations again;

[0050] (5) Repeat steps (3) and (4) until convergence, and finally obtain the structural parameters and process parameters of the single-screw extrusion reactor with optimal performance.

[0051] The relevant data results involved in the iterative process are as follows Figures 3 to 10 As shown, Figure 3 This is the iteration diagram of the Bayesian optimization process. The Bayesian algorithm will predict the structural parameters and process parameters with the best performance based on the initial data, and then transmit these parameters to the CFD simulation software to automatically establish a new screw geometry model and perform simulation calculations. Figure 3 It can be seen that as the iterations proceed, the Bayesian algorithm has found the point with the lowest LDPE outlet mass flow rate or the highest product gasoline outlet mass flow rate in its search history; Figure 4The mass fractions of LDPE at different axial positions under five different operating conditions are shown in Table 1. The lower the mass fraction of LDPE, the more intense the reaction. It can be seen that different operating conditions have a great influence on the catalytic cracking of LDPE. When the pitch is 60 mm, the rotation speed is 100 rpm, and the catalyst volume fraction is 0.45, the LDPE is almost cracked at the reactor outlet. However, when the pitch is 46 mm, the rotation speed is 140 rpm, and the catalyst volume fraction is 0.25, the LDPE is only cracked by 75% at the reactor outlet. Figures 5 to 9 The flow field diagram inside the reactor is shown when the pitch is 60 mm, the rotation speed is 120 rpm, and the catalyst volume fraction is 0.45. Figure 5 and Figure 6 It can be seen that LDPE is not directly cracked into light olefin products at the beginning of the reaction, but combines with the catalyst to form the intermediate product LDPEC. Figure 7 and Figure 9 The gasoline and coke products are formed by the cracking of LDPEC. As the heat transfer between the screw and the fluid increases, the temperature of the fluid increases, and the reaction of LDPE directly cracking into G1 begins. Figure 8 It can be seen that light olefin products begin to form at the axial distance Z = 150 mm and account for 12.1% at the outlet; Figure 10 This is the temperature diagram inside the reactor. The inlet temperature of LDPE is 573K, while the screw temperature is 973K. The fluid temperature at the reactor outlet has reached 850K. The heat transfer between the screw and the fluid makes the reaction more complete.

[0052] Example 2

[0053] A computer device comprising: at least one processor and a memory communicatively connected to the at least one processor;

[0054] The memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the single-screw extrusion reactor design method based on Bayesian optimization as described in Example 1 is implemented.

[0055] Example 3

[0056] A computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the single-screw extrusion reactor design method based on Bayesian optimization as described in Example 1.

[0057] It has been verified that the single-screw extrusion reactor design method based on Bayesian optimization of the present invention provides optimal operating conditions and structural parameters. The pitch, screw speed, catalyst dosage, etc. are all involved in the optimization, which can greatly improve the conversion rate and yield; it is easy to operate and has good consistency, and is easy to promote and apply. The application of the above method can provide a potential path for the design of single-screw extrusion reactors and has good application prospects.

[0058] Those skilled in the art should understand that they can implement variations by combining the prior art with the above embodiments, which will not be described in detail here. Such variations do not affect the essence of the present invention and will not be described in detail here.

[0059] The above describes the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the above-mentioned specific embodiments, and the devices and structures that are not described in detail should be understood to be implemented in a common manner in the art; any technician familiar with the art can use the above-mentioned disclosed methods and technical contents to make many possible changes and modifications to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, or modify them into equivalent embodiments of equivalent changes, which does not affect the essential content of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention that do not depart from the content of the technical solutions of the present invention are still within the scope of protection of the technical solutions of the present invention.

Claims

1. A single-screw extrusion reactor design method based on Bayesian optimization, characterized by: The following steps are involved: (1) Determine the key structural parameters and process parameters of the single-screw extrusion reactor, and establish an initial parametric geometric model of the single-screw extrusion reactor based on these parameters. The key structural parameters and process parameters include screw pitch, screw speed, and catalyst dosage. (2) Importing the initial parameterized geometric model established in step (1) into the simulation software to simulate the LDPE catalytic cracking process and obtain the performance parameters of the reactor model; (3) Implementing a parameter communication interface between the simulation software and the Bayesian optimization algorithm through a script, using the performance parameters obtained in step (2) as the objective function, and automatically optimizing the screw pitch, screw speed, and catalyst dosage using the Bayesian optimization algorithm; (4) Automatically update the parameters of the reactor model according to the optimization results of step (3) and perform simulation calculations again; (5) Repeat steps (3) and (4) until convergence, and finally obtain the structural parameters and process parameters of the single-screw extrusion reactor with optimal performance.

2. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 1, characterized in that: The performance parameters include the mass flow rate of the reactant LDPE, the mass flow rate of the product gasoline and the mass flow rate of the by-product coke at the reactor outlet.

3. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 1, characterized in that: The simulation software is CFD simulation software; The script is a Python script.

4. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 3, characterized in that: The Python script code realizes the automation of the simulation process through the CFD parameter interface.

5. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 1, characterized in that: The simulation software sets the catalytic cracking reaction as a steady-state continuous reaction process.

6. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 3, characterized in that: The initial parameterized geometric model and CFD simulation boundary conditions of the single-screw extrusion reactor are both parameterized designs.

7. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 1, characterized in that: The automatic optimization process can complete the continuous iterative simulation of the reactor model without human intervention.

8. The single-screw extrusion reactor design method based on Bayesian optimization according to claim 1, characterized in that: The catalyst dosage is the catalyst inlet volume fraction.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the single-screw extrusion reactor design method based on Bayesian optimization according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the single-screw extrusion reactor design method based on Bayesian optimization according to any one of claims 1 to 8 is implemented.

Images