DOI：10.3969/j.issn.1672-3872.2024.08.008
作 者：徐 俊1 ，李 强1 ，马盛力1 ，黄 慧1 ，陈 伟1 ，詹 映2
（1. 上海烟草集团有限责任公司上海烟草储运公司，上海 200439；2. 上海创和亿电子科技发展有限公司，上海 200092）
摘 要：【目的】烟叶原料水分含量是影响其品质的主要因素之一，及时掌握烟叶水分变化情况，从而及时采取具有针对性的养护措施，是烟叶原料防霉的基本要求。【方法】采用蒙特卡洛偏最小二乘交叉检验法剔除异常样本后，用偏最小二乘法（PLS）建立了单部位烟叶、多部位烟叶以及不同产地烟叶的水分校正模型，并对模型进行了验证。【结果】单部位烟叶含水率构建的模型相关系数为0.79，平均绝对误差为0.24，相对误差为1.99%；多部位烟叶含水率构建的模型相关系数为0.81，平均绝对误差为0.30，相对误差为2.44%；多产地烟叶含水率构建的模型相关系数为0.88，平均绝对误差为0.27，相对误差为2.24%。因此，无论是用单产地单部位烟叶还是用单产地多部位烟叶，或者是用多个产地的烟叶进行建模，所建立的模型的精度和稳定性都能够满足实际应用的需求。【结论】微波水分速测技术所建立的水分模型能够在烟叶贮存养护中广泛应用，便携式微波水分仪可有效地在不改变垛型、不移动烟箱、不破坏（打开）烟箱的情况下，实现烟垛棱边烟箱内烟叶含水率的快速准确在线检测，并减少人工测定所引起的误差，大大缩短检测时间。
关键词：烟叶；仓储；水分检测；快速检测技术；偏最小二乘法
Application of Portable Microwave Moisture Meter in Moisture Content Detection of Tobacco Leaves in Storage
Xu Jun1 , Li Qiang1 , Ma Shengli1 , Huang Hui1 , Chen Wei1 , Zhan Ying2
(1.Shanghai Tobacco Group Co., Ltd., Shanghai Tobacco Storage and Transportation Company, Shanghai 200439; 2.Shanghai Micro Vision Technology LTD, Shanghai 200092)
Abstract: [Objective] The moisture content of tobacco raw materials is one of the main factors affecting its quality. Timely grasp of the tobacco leaf moisture changes, so as to take timely and targeted maintenance measures, is the basic requirement for mold prevention of tobacco raw materials. [Method] After eliminating the abnormal samples by Monte Carlo partial least squares cross test, moisture correction models for single part tobacco, multi part tobacco, and tobacco from different regions were established using the partial least squares (PLS) method, and the models were validated. [Result] The correlation coefficient of the model constructed for the moisture content of single part tobacco leaves is 0.79, with an average absolute error of 0.24 and a relative error of 1.99%. The correlation coefficient of the model constructed for the moisture content of multiple tobacco leaves is 0.81, with an average absolute error of 0.30 and a relative error of 2.44%. The correlation coefficient of the model constructed for the moisture content of tobacco leaves from multiple regions is 0.88, with an average absolute error of 0.27 and a relative error of 2.24%. Therefore, whether it is modeled with a single part of tobacco from a single place of production or with multiple parts of tobacco from a single place of production, or with tobacco from multiple places of origin, the accuracy and stability of the established model can meet the needs of practical applications. [Conclusion] The moisture model established by microwave moisture rapid measurement technology can be widely used in tobacco storage and maintenance. The portable microwave moisture meter can effectively achieve fast and accurate on-line detection of the moisture content of tobacco leaves in the tobacco box at the edge of the stack without changing the stack type, moving the tobacco box and destroying (opening) the tobacco box, and reduce errors caused by manual measurement, greatly shortening the detection time.
Keywords: tobacco leaves; storage; moisture detection; rapid detection technology; partial least squares method
引文信息：[1]徐俊，李强，马盛力，等.便携式微波水分仪在仓储烟叶含水率检测中的应用[J].南方农机，2024，55(8)：27-32+56.
查看全文请下载PDF文件↓
