Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Cotton Ginning Research » Research » Publications at this Location » Publication #399417
Research Project: Development and Evaluation of Novel Technologies to Improve Fiber Quality and Increase Profitability in Cotton Processing

Location: Cotton Ginning Research

Title: Inline real-time moisture sensing system for gin cotton

Author
item TO, FILIP - Mississippi State University
item GAY, LUCAS - Mississippi State University
item Thomas, Joseph
item Donohoe, Sean

Submitted to: National Cotton Council Beltwide Cotton Conference
Publication Type: Proceedings
Publication Acceptance Date: 11/4/2022
Publication Date: 6/1/2023
Citation: To, F., Gay, L., Thomas, J.W., Donohoe, S.P. 2023. Inline real-time moisture sensing system for gin cotton. National Cotton Council Beltwide Cotton Conference, January 10-12, 2023, New Orleans, LA. P.86-94. 2023.

Interpretive Summary:

Technical Abstract: The objective of this study is to assess the performance of a select sensor in measuring moisture content of cotton within the moisture range of 6%-13%. It is the first step in developing a system to monitor moisture in real time in the ginning environment with a long-term goal of developing a closed loop system in which cotton moisture may be maintained optimally throughout the ginning. The assessment was carried out in two phases: static (cotton velocity = 0) and dynamic (in-stream). During the static testing phase, models to predict moisture content were created and a 95% Confidence Interval of [-0.619,0.619], Percent Residual Accuracy of 97.586%, and a mean error of 4.396% were achieved. The tests were done with 110 samples. The Bland-Altman plot yielded a curve with a slight positive trend, suggesting that the data could be modeled more appropriately by multiple models with each predicting a narrow range of moisture contents. The dynamic testing phase saw similar results as the static tests in overall trends, however the predicted values tended to be less accurate than in static testing. Static testing had a RMSE of 0.313 compared to the dynamic testing having a RMSE of 0.737 [MC%]. This may stem from the relatively low number of data points collected during the dynamic test as compared to the amount of data collected in the static phase.