(109f) Statistics Pattern Analysis-Based Multiclass Woodchip Moisture Classification System Using Short-Range Iot Wi-Fi

The pulp and paper industry is the third-largest consumer of energy in the US industrial sector thereby leading to tremendous opportunities to improve its energy efficiency and productivity. The pulping process, which converts woodchips into pulp by displacing lignin from cellulose fibers, is one of the most important operations in a pulp and paper mill. For the pulping process in a pulp & paper plant that uses wood as raw material, it is important to have real-time knowledge about the moisture content of the woodchips so that the process can be optimized and/or controlled correspondingly to achieve satisfactory product quality while minimizing the consumption of energy and chemicals. Both destructive and non-destructive methods have been developed for estimating moisture content in woodchips, but these methods are often lab-based that cannot be implemented online due to time constraint [1], relatively complex and costly to install and operate[2], or too fragile to stand the harsh manufacturing environment. Due to these limitations, at present, the moisture level is measured only a few times throughout the year in the pulp and paper industry.

To address these limitations, we propose a non-destructive and economic approach based on the Internet-of-Things (IoT) based Wi-Fi and use channel state information (CSI) to estimate the moisture content in woodchips. We extract CSI by modifying the open-source device drivers for Intel Wi-Fi link 5300 Network Interface Card based on CSITool [3]. CSI contains information about the channel in the form of individual data sub-carriers capturing indoor channel features like the effect of scattering, fading, and power decay with distance. We hypothesize that when a Wi-Fi signal passes through wood chips, the water or moisture content in woodchips would cause a corresponding change in the Wi-Fi signal which is then captured in CSI. The goal is to develop a data-driven model to extract the relationship between the change in CSI and the woodchip moisture content. CSI has been successfully used for moisture detection in wheat [4]. However, compared to wheat moisture detection, woodchip moisture detection is much more challenging due to the significant heterogeneity in the shape, structure and arrangement of the woodchips. For example, the woodchip arrangement in the container is expected to have a significant impact on the CSI data.

In this work, our goal is to develop a robust approach that is insensitive to the heterogeneity of the woodchips, and in particular the effect of shuffling, to achieve a consistent and accurate moisture estimation. To achieve this goal, our proposed approach is based on statistics pattern analysis (SPA) framework [5][6] we developed previously, where SPA is integrated with different classification approaches to build robust multivariate statistical models for multiclass moisture classification in woodchips. Specifically, SPA is used to extract features from extremely noisy raw Wi-Fi CSI data thereby converting the messy big data into so called smart data for accurate sensing. These features are then used to build multiclass classification models using machine learning techniques including linear discriminant analysis (LDA), subspace discriminant classification (SDC), support vector machines (SVM), and artificial neural networks (ANN). We demonstrate an extensive study on the above-mentioned approaches for multiclass classification which, when combined with SPA, provide potential solutions for estimating woodchip moisture content for the pulp and paper industry. Our results show highly accurate classification of more than 20 moisture levels even when two moisture classes are separated by a very small margin of <1% (wet basis). Our approach is not only economic and non-destructive but can also be implemented online, making it highly desirable for the pulp and paper manufacturing environment.

[1] R. Govett, “A PRACTICAL GUIDE FOR THE DETERMINATION OF MOISTURE CONTENT OF WOODY BIOMASS A Practical Handbook of Basic Information, Definitions, Calculations, Practices and Procedures for Purchasers and Suppliers of Woody Biomass.”

[2] S. O. F. Sensing, W. H. O. Uses, and E. Method, “Learn the Six Methods For Determining Moisture.”

[3] D. Halperin, W. Hu, A. Sheth, and D. Wetherall, “Tool Release: Gathering 802.11n Traces with Channel State Information,” ACM SIGCOMM Comput. Commun. Rev., 2011.

[4] W. Yang, X. Wang, S. Cao, H. Wang, and S. Mao, “Multi-class wheat moisture detection with 5GHz Wi-Fi: A deep LSTM approach,” in Proceedings - International Conference on Computer Communications and Networks, ICCCN, 2018.

[5] Q. P. He and J. Wang, “Statistics pattern analysis: A new process monitoring framework and its application to semiconductor batch processes,” AIChE J., 2011.

[6] J. Wang and Q. P. He, “Multivariate Statistical Process Monitoring Based on Statistics Pattern Analysis,” Ind. Eng. Chem. Res., 2010.