Article: Early warning system of cigarette process quality combined with intelligent sensing technology Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.23 pp.1 - 23 Abstract: In this paper, an early warning system of cigarette process quality combined with intelligent sensing technology is proposed to improve the quality of cigarette process production. A PCA multi-block modelling algorithm based on autoencoder feature extraction is proposed to extract autoencoder features from each sub-block, and the statistics of all sub-blocks are fused by Bayesian inference to make the monitoring results more intuitive. Compared with the traditional PCA and AE-PCA detection methods, the AE-MPCA algorithm proposed in this paper improves the abnormality detection accuracy of the drum leaf drying production process, and realises the accurate alarm of quality abnormalities, thus providing technical support for the early warning of subsequent cigarette process quality. In the subsequent process of cigarette process quality control, the application scope of intelligent sensing technology can be further improved to promote the effect of cigarette process quality control. Inderscience Publishers - linking academia, business and industry through research

Title: Early warning system of cigarette process quality combined with intelligent sensing technology

Authors: Rongya Zhang; Mingchang Liu; Yaping Ma; Tingjie Bao; Peng Kuang; Wenbin Feng; Jiaxi Ni; Lingying He; Bin Yang; Chao Yang; Wu Wen; Ruifang Gu

Addresses: Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Shifang Cigarette Factory, China Tobacco Sichuan Industrial Co., Ltd., No. 158, South Section of Yinghuashan Road, Shifang, Sichuan 618400, China ' Shifang Cigarette Factory, China Tobacco Sichuan Industrial Co., Ltd., No. 158, South Section of Yinghuashan Road, Shifang, Sichuan 618400, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Shifang Cigarette Factory, China Tobacco Sichuan Industrial Co., Ltd., No. 158, South Section of Yinghuashan Road, Shifang, Sichuan 618400, China ' Shifang Cigarette Factory, China Tobacco Sichuan Industrial Co., Ltd., No. 158, South Section of Yinghuashan Road, Shifang, Sichuan 618400, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China ' Technology Center, China Tobacco Sichuan Industrial Co., Ltd., Section 2, Chenglong Avenue, Longquanyi District, Chengdu, Sichuan 610100, China

Abstract: In this paper, an early warning system of cigarette process quality combined with intelligent sensing technology is proposed to improve the quality of cigarette process production. A PCA multi-block modelling algorithm based on autoencoder feature extraction is proposed to extract autoencoder features from each sub-block, and the statistics of all sub-blocks are fused by Bayesian inference to make the monitoring results more intuitive. Compared with the traditional PCA and AE-PCA detection methods, the AE-MPCA algorithm proposed in this paper improves the abnormality detection accuracy of the drum leaf drying production process, and realises the accurate alarm of quality abnormalities, thus providing technical support for the early warning of subsequent cigarette process quality. In the subsequent process of cigarette process quality control, the application scope of intelligent sensing technology can be further improved to promote the effect of cigarette process quality control.

Keywords: intelligent perception; cigarettes; process quality; early warning.

DOI: 10.1504/IJICT.2025.147127

International Journal of Information and Communication Technology, 2025 Vol.26 No.23, pp.1 - 23

Received: 19 Feb 2025
Accepted: 11 May 2025
Published online: 10 Jul 2025 *

Title: Early warning system of cigarette process quality combined with intelligent sensing technology

Keep up-to-date