文献类型：期刊文献

英文题名：A novel detection method of fine particle concentration based on lab-on-a-chip with micro-electric field and image recognition of artificial intelligence

作者：Lei, Donghui[1];Huang, Zhenpeng[1];Chen, Jitong[1];Yang, Xiaoyong[1];Yin, Wei[1];Bai, Zhishan[1];Zhu, Yong[1]

机构：[1]East China Univ Sci & Technol, Sch Mech & Power Engn, Shanghai 200237, Peoples R China

年份：2025

卷号：13

期号：6

外文期刊名：JOURNAL OF ENVIRONMENTAL CHEMICAL ENGINEERING

收录：;EI(收录号：20254519468984);WOS:【SCI-EXPANDED(收录号:WOS:001584160300013)】；

基金：This work was supported by the National Natural Science Foundation of China (No. 22578129) .

语种：英文

外文关键词：Artificial intelligence; Lab-on-a-chip; On-line detection; Fine particle; Image recognition

摘要：In this work, a novel detection method of fine particle concentration is developed, based on a lab-on-a-chip featuring a micro-electric field and image recognition using artificial intelligence. Utilizing the lab-on-a-chip, fine particles can be directly captured from continuous gas-solid flow by the micro-electric field. Concurrently, an object detection algorithm based on machine learning is employed to automatically recognize the deposited particle agglomerates and analyze their length. Real-time optical images of the particles are acquired by an sCMOS camera and transmitted to the object detection model for feature extraction. Distribution of the micro-electric field is simulated based on Maxwell's equation to better understand the particle dynamics. Considering both the electric field and diffusion charging mechanisms, particle charging number is obtained, and the particle migration is observed via high-speed imaging. Furthermore, based on theoretical analysis of particle dynamics, the relationship between the features extracted from optical images and the operational parameters is established, mainly considering the applied voltage. An optimal voltage detection window for monitoring particle concentration is determined through the adjustment of the electric field strength. Overall, this method offers great potential for advancing particle imaging and monitoring, providing valuable insights for various applications in chemical engineering.