View in EDS

Prediction of deep learning algorithms for the microemulsion generation conditions in complex environments.

Saved in:
Bibliographic Details
Title: Prediction of deep learning algorithms for the microemulsion generation conditions in complex environments.
Authors: Li, Hao1 (AUTHOR), Ding, Tianshun1 (AUTHOR), Tao, Shengyang1 (AUTHOR) taosy@dlut.edu.cn
Source: Journal of the Taiwan Institute of Chemical Engineers. Feb2026, Vol. 179, pN.PAG-N.PAG. 1p.
Subjects: Microemulsions, Microfluidics, Prediction models, Machine learning, Deep learning, Fluid dynamics, Convolutional neural networks
Abstract: • RAM-CNN achieves 100 % fluid and 95.8 % droplet morphology recognition accuracy. • GAN-LGBMnet enhances small-sample learning with MAPE <7 % for droplet prediction. • User-friendly GUI enables ≤2-click workflows for non-expert microfluidic design. • MO-DNN predicts single/double emulsion parameters with <10 % error on unseen data. Constructing high-quality datasets for AI-driven microemulsion prediction remains challenging due to limitations in chip design, imaging hardware, and fluid dynamics. Conventional semi-empirical models suffer from poor accuracy and generalizability in complex nonlinear systems, while traditional microfluidic chips often yield non-spherical droplets. Optical constraints in capillary-based systems further hinder data acquisition. A deep learning framework integrates three components: (1) A multi-branch CNN with residual modules and self-attention (RAM-CNN) for robust droplet/fluid morphology recognition; (2) GAN-LGBMnet, combining adversarial networks and LightGBM, to augment small datasets and analyze key features; (3) A multi-output neural network (MO-DNN) predicting single/double-emulsion parameters. The system is deployed via an intuitive GUI (MICA) for non-specialist use. RAM-CNN achieves 100 % fluid-regime and 95.8 % droplet-morphology recognition accuracy, maintaining >91 % performance under ±28 % brightness variations. Enhanced by GAN-LGBMnet, MO-DNN predicts droplet diameters and generation rates with MAPE <7 % for both emulsion types. The MICA platform demonstrates <10 % error on unseen data, enabling precise emulsion design. This work bridges theoretical models with practical microfluidic optimization through automated, user-friendly AI tools. [Display omitted] [ABSTRACT FROM AUTHOR]
Copyright of Journal of the Taiwan Institute of Chemical Engineers is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:• RAM-CNN achieves 100 % fluid and 95.8 % droplet morphology recognition accuracy. • GAN-LGBMnet enhances small-sample learning with MAPE <7 % for droplet prediction. • User-friendly GUI enables ≤2-click workflows for non-expert microfluidic design. • MO-DNN predicts single/double emulsion parameters with <10 % error on unseen data. Constructing high-quality datasets for AI-driven microemulsion prediction remains challenging due to limitations in chip design, imaging hardware, and fluid dynamics. Conventional semi-empirical models suffer from poor accuracy and generalizability in complex nonlinear systems, while traditional microfluidic chips often yield non-spherical droplets. Optical constraints in capillary-based systems further hinder data acquisition. A deep learning framework integrates three components: (1) A multi-branch CNN with residual modules and self-attention (RAM-CNN) for robust droplet/fluid morphology recognition; (2) GAN-LGBMnet, combining adversarial networks and LightGBM, to augment small datasets and analyze key features; (3) A multi-output neural network (MO-DNN) predicting single/double-emulsion parameters. The system is deployed via an intuitive GUI (MICA) for non-specialist use. RAM-CNN achieves 100 % fluid-regime and 95.8 % droplet-morphology recognition accuracy, maintaining >91 % performance under ±28 % brightness variations. Enhanced by GAN-LGBMnet, MO-DNN predicts droplet diameters and generation rates with MAPE <7 % for both emulsion types. The MICA platform demonstrates <10 % error on unseen data, enabling precise emulsion design. This work bridges theoretical models with practical microfluidic optimization through automated, user-friendly AI tools. [Display omitted] [ABSTRACT FROM AUTHOR]
ISSN:18761070
DOI:10.1016/j.jtice.2025.106414