Online reconstruction method for chaotic attractors in stirred reactor
By combining fluid mechanics coupled discrete particle simulation and phase space reconstruction technology with deep learning, the chaotic attractors in the stirred reactor are predicted in real time, which solves the shortcomings of traditional methods in the chemical fluid mixing process and improves the reactor Efficiency and uniformity of material transmission.
Patent Information
- Application Number
- CN202410945121.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-07-15
AI Technical Summary
Traditional methods are difficult to accurately understand and predict the dynamic behavior of the fluid in real time during the chemical fluid mixing process, resulting in insufficient material transfer efficiency and uniformity in the reactor.
Use computational fluid mechanics coupled discrete particles to collect the motion data of pulsed particles in the stirred reactor, use phase space reconstruction technology to obtain chaotic attractors, and use trained long and short-term memory network model to predict chaotic attractors to achieve future periods Real-time prediction of dynamic behavior.
Real-time prediction of chaotic attractors in the stirred reactor is achieved, the optimization and intelligent development of the chemical fluid mixing process are improved, and new technical support is provided for reactor design.
Smart Images

Figure CN119047282A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of chaotic mixing of stirred fluids and the field of machine learning, and in particular to an online reconstruction method of a chaotic attractor in a stirred reactor. Background Art
[0002] In the process of chemical fluid mixing, the complexity of the flow field structure directly affects the efficiency and uniformity of material transfer in the reactor. Traditional methods usually rely on numerical simulation or experimental measurement to understand and predict fluid dynamic behavior, but these methods are limited in real-time and accuracy. Summary of the invention
[0003] The object of the present invention is to provide a method for online reconstruction of a chaotic attractor in a stirred reactor, comprising the following steps:
[0004] 1) Perform computational fluid dynamics coupled discrete particle simulation on a stirred reactor filled with materials, and collect pulse line particle motion data in the stirred reactor;
[0005] 2) Processing the pulse particle motion data in the stirred reactor using phase space reconstruction to obtain the chaotic attractor in the stirred reactor;
[0006] 3) The chaotic attractor prediction model is used to process the chaotic attractor and obtain the prediction result of the chaotic attractor in the stirred reactor in the future t period.
[0007] Furthermore, the flow characteristics of the material inside the stirred reactor belong to laminar flow or transitional flow.
[0008] Furthermore, the stirring speed of the stirred reactor is 0 to 600 rpm.
[0009] Furthermore, the computational fluid dynamics coupled discrete particle simulation refers to using Fluent software to simulate and obtain the change of the position of the tracer particles in the stirred reactor over time.
[0010] Furthermore, the cross-sectional shape of the stirred reactor includes but is not limited to square, rectangular, circular, and elliptical.
[0011] Furthermore, the materials contained in the stirred reactor include high molecular polymers, soft materials, high-end grease, and glycerin.
[0012] Furthermore, the pulse line particle motion data refers to the spatial position of particles at different times at the same injection position in the stirred reactor.
[0013] For a cylindrical stirred reactor with a diameter of T=200 mm, the initial injection position includes but is not limited to (X, Y, Z)=(-60 mm, 0 mm, 60 mm).
[0014] Furthermore, the phase space reconstruction technique is used to observe the trajectory distribution of the pulse line particles moving in the X and Y directions.
[0015] Furthermore, the chaotic attractor prediction model is a trained long short-term memory network model. Model parameters: input layer number input_size=1, hidden layer number hidden_size=50, hidden layer number num_layers=2, output layer number output_size=1, learning rate lr=0.01, iteration number num_epochs=200.
[0016] The technical effect of the present invention is unquestionable. The present invention proposes an online reconstruction method based on chaotic attractors, which generates chaotic attractors in stirred reactors by coupling data collected by discrete particle simulation with computational fluid dynamics, and using phase space reconstruction technology. Combined with deep learning prediction algorithms, this method realizes real-time prediction of chaotic attractors, providing new technical means and theoretical support for the optimization and intelligent development of chemical fluid mixing processes. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 is the training data prediction and loss function.
[0018] Figure 2 is the test data prediction.
[0019] Figure 3 It is the chaotic attractor prediction diagram. DETAILED DESCRIPTION
[0020] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.
[0021] Embodiment 1:
[0022] See also Figures 1 to 3 , a method for online reconstruction of a chaotic attractor in a stirred reactor, comprising the following steps:
[0023] 1) Perform computational fluid dynamics coupled discrete particle simulation on a stirred reactor filled with materials, and collect pulse line particle motion data in the stirred reactor;
[0024] 2) Processing the pulse particle motion data in the stirred reactor using phase space reconstruction to obtain the chaotic attractor in the stirred reactor;
[0025] 3) The chaotic attractor prediction model is used to process the chaotic attractor and obtain the prediction result of the chaotic attractor in the stirred reactor in the future t period.
[0026] The flow characteristics of the material inside the stirred reactor belong to laminar flow or transition flow.
[0027] The stirring speed of the stirred reactor is 0 to 600 rpm.
[0028] The computational fluid dynamics coupled discrete particle simulation refers to using Fluent software to simulate and obtain the change of the position of the tracer particles in the stirred reactor over time. The cross-sectional shape of the stirred reactor includes but is not limited to square, rectangle, circle, and ellipse.
[0029] The materials contained in the stirred reactor include high molecular polymers, soft materials, high-end greases, and glycerin.
[0030] The pulse line particle motion data refers to the spatial position of particles at different times at the same injection position in the stirred reactor.
[0031] For a cylindrical stirred reactor with a diameter of T=200 mm, the initial injection position includes but is not limited to (X, Y, Z)=(-60 mm, 0 mm, 60 mm).
[0032] The phase space reconstruction technique is used to observe the trajectory distribution of the pulse line particles moving in the X and Y directions.
[0033] The chaotic attractor prediction model is a trained long short-term memory network model. Model parameters: input layer number input_size=1, hidden layer number hidden_size=50, hidden layer number num_layers=2, output layer number output_size=1, learning rate lr=0.01, iteration number num_epochs=200.
[0034] Embodiment 2:
[0035] A method for online reconstruction of a chaotic attractor in a stirred reactor comprises the following steps:
[0036] 1) Perform computational fluid dynamics coupled discrete particle simulation on a stirred reactor filled with materials, and collect pulse line particle motion data in the stirred reactor;
[0037] 2) Using phase space reconstruction technology to process the pulse particle motion data in the stirred reactor, and obtain the chaotic attractor in the stirred reactor;
[0038] 3) The chaotic attractor prediction model is used to process the chaotic attractor and obtain the prediction result of the chaotic attractor in the stirred reactor in the future t period.
[0039] Embodiment 3:
[0040] A method for online reconstruction of chaotic attractors in a stirred reactor, the technical content is the same as that of Example 2, further, the computational fluid dynamics coupled discrete particle simulation refers to using Fluent software to simulate and obtain the change of the position of the tracer particles in the stirred reactor over time. For a cylindrical stirred reactor with a diameter of T = 200 mm, the initial injection position includes but is not limited to (X, Y, Z) = (-60 mm, 0 mm, 60 mm).
[0041] Embodiment 4:
[0042] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of Examples 2-3, and further, the flow characteristics of the material inside the stirred reactor belong to laminar flow or transitional flow.
[0043] Embodiment 5:
[0044] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of embodiments 2-4, and further, when the stirred reactor is stirred, the stirring speed is 0 to 600 rpm.
[0045] Embodiment 6:
[0046] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of Examples 2-5, and further, the cross-sectional shape of the stirred reactor includes but is not limited to square, rectangular, circular, and elliptical.
[0047] Embodiment 7:
[0048] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of Examples 2-6, and further, the materials contained in the stirred reactor include high molecular polymers, soft materials, high-end grease, and glycerin.
[0049] Embodiment 8:
[0050] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of Examples 2-7, and further, the pulse line particle motion data refers to the spatial position of particles at different times at the same injection position in the stirred reactor. For a cylindrical stirred reactor with a diameter of T=200 mm, the same injection position includes but is not limited to (X, Y, Z)=(-60 mm, 0 mm, 60 mm).
[0051] Embodiment 9:
[0052] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of embodiments 2-8, and further, the phase space reconstruction technology mainly refers to examining the trajectory distribution of pulse line particles moving in the X and Y directions.
[0053] Embodiment 10:
[0054] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of embodiments 2-9, and further, the chaotic attractor prediction model is a trained long short-term memory network model. The input data is the spatial distribution data of the pulse line particles from 0 to t, and the output is the spatial distribution data of the pulse line particles from t to t+m.
[0055] Embodiment 11:
[0056] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of embodiments 2-10, and further, the long short-term memory network model parameters: the number of input layer layers input_size=1, the number of hidden layer layers hidden_size=50, the number of hidden layer layers num_layers=2, the number of output layer layers output_size=1, the learning rate lr=0.01, and the number of iterations num_epochs=200.
[0057] Embodiment 12:
[0058] A method for online reconstruction of a chaotic attractor in a stirred reactor, the technical content of which is the same as any one of embodiments 1-11, and further, an example of the particle motion data is shown in the following table:
[0059]
[0060]
[0061] [Note] The more data in the flow data table, the better. In actual operation, at least 120,000 records are required.
[0062] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.
[0063] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0064] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0066] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.
[0067] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A method for online reconstruction of chaotic attractors in a stirred reactor. It is characterized in that The following steps are involved: 1) Perform computational fluid dynamics coupled discrete particle simulation on a stirred reactor filled with materials, and collect pulse line particle motion data in the stirred reactor; 2) Processing the pulse particle motion data in the stirred reactor using phase space reconstruction to obtain the chaotic attractor in the stirred reactor; 3) The chaotic attractor prediction model is used to process the chaotic attractor and obtain the prediction result of the chaotic attractor in the stirred reactor in the future t period.
2. The method for online reconstruction of a chaotic attractor in a stirred reactor according to claim 1, It is characterized in that The flow characteristics of the material inside the stirred reactor belong to laminar flow or transition flow.
3. The method for online reconstruction of a chaotic attractor in a stirred reactor according to claim 2, It is characterized in that The stirring speed of the stirred reactor is 0 to 600 rpm.
4. The method for online reconstruction of a chaotic attractor in a stirred reactor according to claim 1, It is characterized in that The computational fluid dynamics coupled discrete particle simulation refers to using Fluent software to simulate and obtain the change of the position of the tracer particles in the stirred reactor over time.
5. The method for online reconstruction of chaotic attractors in a stirred reactor according to claim 1, It is characterized in that The cross-sectional shape of the stirred reactor includes, but is not limited to, square, rectangular, circular, and elliptical.
6. The method for online reconstruction of chaotic attractors in a stirred reactor according to claim 1, It is characterized in that The materials contained in the stirred reactor include high molecular polymers, soft materials, high-end greases, and glycerin.
7. The method for online reconstruction of chaotic attractors in a stirred reactor according to claim 1, It is characterized in that The pulse line particle motion data refers to the spatial position of particles at different times at the same injection position in the stirred reactor.
8. The method for online reconstruction of chaotic attractors in a stirred reactor according to claim 1, It is characterized in that The phase space reconstruction technique is used to observe the trajectory distribution of the pulse line particles moving in the X and Y directions.
9. The method for online reconstruction of a chaotic attractor in a stirred reactor according to claim 1, It is characterized in that The chaotic attractor prediction model is a trained long short-term memory network model. Model parameters: input layer number input_size=1, hidden layer number hidden_size=50, hidden layer number num_layers=2, output layer number output_size=1, learning rate lr=0.01, iteration number num_epochs=200.
Citation Information
Patent Citations
Chaotic phase space optimization reconstruction photovoltaic power generation neural network prediction method
CN109255728A
Chaotic time sequence prediction method based on attention mechanism deep learning
CN111260124A
Method for intelligent model selection and operation parameter intelligent control of stirring reactor and intelligent terminal equipment
CN118286990A
Quantum mechanics instruction production systems, methods, and applications thereof
US20220164501A1
