Simulation and prediction of three-dimensional rotating flows based on convolutional neural networks

Gao, Feng; Zhang, Zhuang; Jia, Chenyang; Zhu, Yin; Zhou, Chunli; Wang, Jingtao

doi:10.1063/5.0113030

Physics of Fluids

2022

DOI: 10.1063/5.0113030

|View full text |Cite

Simulation and prediction of three-dimensional rotating flows based on convolutional neural networks

Feng Gao

Zhuang Zhang

Chenyang Jia

et al.

Abstract: Two deep learning models to reconstruct 3-dimensional (3D) steady-state rotating flows are proposed to capture the spatial information: the 3D Convolutional Encode-decoder and the 3D Convolutional Long Short-term Memory (LSTM) Model. They are based on deep learning methods such as the encoder-decoder convolutional neural network (ED-CNN) and recurrent neural network (RNN). Their common component sare an encoder, a middle layer, and a decoder. The rotating flows in a stirred tank with four inclined blades are c… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2023

2024

Publication Types

Select...

Article3

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

PEDM: A 3D Position Encoding Diffusion Model for Industrial Cracking Furnace Scalar Field Data Generation

Ding,

Cheng,

2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Traditional computational fluid dynamics methods are time-consuming to obtain steady-state scalar fields by iteratively solving partial differential equations. Therefore, many scholars have adopted deep learning methods to construct data-driven surrogate models for more efficient generation of scalar fields. However, unlike the common image data usually used for deep learning modeling, the data used for the surrogate model are results of the computational fluid dynamics simulation models, which have irregular grids. Therefore, inputting the grid data into the deep learning model directly will cause difficulties in model training due to the unpreserved spatial location relationships. Auxiliary grids are traditionally used to pixelize the original irregular grid data and train the model. However, the resolution of the auxiliary grids is difficult to decide, which will result in either the information loss or the increase of computational complexity. In this paper, a 3D position encoding diffusion model (PEDM) is proposed to address the problem of the loss of spatial positional relationships in the grid data, by introducing the grid position encoding. The model uses U-net to learn feature information from the field data and utilizes spatial location within the original grids to encode the position of noised data through MultiLayer Perceptron, providing location information based on the global receptive field of the attention mechanism. This allows the network to learn the distribution of the field data better. Furthermore, the PEDM method is applied to an industrial cracking furnace for field data generation. The average normalized mean absolute error for the field data in the test data set is 0.1295%, indicating sufficient prediction accuracy. The method provides a promising avenue for generating 3D field data in industrial applications, facilitating visualization of reaction processes, virtual sensor deployment, and exploration of design and optimization spaces.

show abstract

PEDM: A 3D Position Encoding Diffusion Model for Industrial Cracking Furnace Scalar Field Data Generation

Ding,

Cheng,

2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

Optimization of painting efficiency applying unique techniques of high-voltage conductors and nitrotherm spray: Developing deep learning models using computational fluid dynamics dataset

2023

View full text Add to dashboard Cite

The impetus of the current three-dimensional Eulerian–Lagrangian work is to analyze the impact of simultaneously using the inventive high-voltage conductors and Nitrotherm spraying technique for maximizing the industrial painting process efficiency. This investigation employs high-fidelity computational fluid dynamics (CFD) results in deep learning models as an input dataset. The novel conductors are called high-voltage retractable blades (HVRB) and high-voltage adjustable control-ring (HVACR) mounted on the head of the electrostatic rotating bell sprayer. The influence of dominant operational parameters, such as temperature and velocity of injected nitrogen or air, droplets' electric charge values, and their size ranges, and electric field density are examined in the considered database for the Nitrotherm spraying methodology. This broad range of parametric investigation illustrates that the inclusion of shaping nitrogen flow, manipulated electric field density, and droplet charging weights significantly affect the spraying deposition rate. The pressurized clean heated nitrogen flow, which is injected from the nozzles of the atomizers, positively redirects and harmonizes the charged droplets that construct an optimized spray plume pattern with a smaller diameter. Using innovative HVRB and HVACR conductors is manipulated the electric fields and leads to denser distribution, intensifying the acting electric force on the droplets, resulting in higher spraying transfer efficiency (TE) and thicker film formation. Based on the results, employing the introduced conductors in combination with the heated nitrogen instead of air leads to higher TE, rare overspray occurrence, formation of an esthetic paint film, lower paint consumption, and application time. Also, the collected complete database is employed for machine learning investigation to predict flow with high accuracy, aiming to reduce computational time/cost. A convolutional auto-encoder is used to reduce the computational cost with just 10% of the initial CFD computations, with a mean error of 1% on the prediction of the deposited droplet areas of the spray. The analysis revealed that by employing recurrent convolutional layers, superior capturing of the input pattern is obtained, which significantly aids the final prediction.

show abstract

A deep-learning-based compact method for accelerating the electrowetting lattice Boltzmann simulations

Zhuang,

Xu,

Zeng

et al. 2024

Physics of Fluids

View full text Add to dashboard Cite

Research on the electrowetting of micro- and nanoscale droplets is essential for microfluidics and nanomaterials applications. A lattice-Boltzmann-electrostatics (LBES) method is an effective and accurate method for simulating this process. However, the electric potential field in each time step requires numerous iterative calculations to converge. Therefore, there is a trade-off dilemma between using high-density lattice fields to improve simulation refinement and low-density lattice fields to reduce computing costs in simulations. Fortunately, deep learning techniques can enhance the computing efficiency of electric potential fields, providing an efficient and accurate solution for electrowetting studies in fine-grained fields. In this study, a compact LBES (C-LBES), a computationally accelerated model for an electric potential field with spatiotemporal prediction capability, is developed by combining the advantages of a recurrent residual convolutional unit and a convolutional long-short-term memory unit. A loss function incorporating a geometric boundary constraint term and a self-cyclic prediction scheme are introduced according to the characteristics of the prediction task, which further improves the prediction accuracy of the model and the computing efficiency of the electric potential field. The model is validated with small datasets, and the results show that the C-LBES model with the self-cyclic prediction scheme improves the computing efficiency of the conventional LBES method by a factor of 10 and provides high-precision results when predicting a two-dimensional convergent electric potential field with a lattice size of (110, 160). In the generalization experiments, the average absolute error of the calculated results remains in the same order of magnitude as the accuracy experimental results.

show abstract

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

hi@scite.ai

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Made with 💙 for researchers

Part of the Research Solutions Family.

Simulation and prediction of three-dimensional rotating flows based on convolutional neural networks

Cited by 3 publications

References 41 publications

PEDM: A 3D Position Encoding Diffusion Model for Industrial Cracking Furnace Scalar Field Data Generation

PEDM: A 3D Position Encoding Diffusion Model for Industrial Cracking Furnace Scalar Field Data Generation

Optimization of painting efficiency applying unique techniques of high-voltage conductors and nitrotherm spray: Developing deep learning models using computational fluid dynamics dataset

A deep-learning-based compact method for accelerating the electrowetting lattice Boltzmann simulations

Contact Info

Product

Resources

About