Scholarlink

<p>This paper discusses ways to improve the accuracy of measuring moisture content in bulk materials, which is critical for ensuring product quality, preservation, and process efficiency in agriculture, food industry, and construction. A&#xa0;brief overview of moisture content measurement techniques is provided, showing that conventional methods, including gravimetric analysis, while highly accurate, lack the ability to provide real-time and continuous monitoring. The paper introduces a&#xa0;newly developed method for measuring moisture content, which utilizes a&#xa0;measurement system that includes the sensor and intellectual components. The sensor component represents a&#xa0;moisture meter comprising capacitive sensors to measure the dielectric constant of the material. The signals from the sensors are transmitted to a&#xa0;data acquisition and pre-processing unit for filtering and normalization. The sensor component provides stable measurements of moisture content for wheat, corn, and sand in the range of 6–25%. The intelligent component of the system includes a&#xa0;regression model, which accounts for the effects of dielectric constant, bulk material density, and ambient temperature on the accuracy of moisture content measurements. It represents a&#xa0;multi-parameter linear regression model implemented using the scikit-learn library (Python). A&#xa0;10-fold cross-validation was used to assess the stability of the model. Experimental studies showed an average absolute measurement error of less than 1.8% and a&#xa0;determination coefficient of more than 0.89, thus confirming the stability and reproducibility of the system. The presented approach demonstrates that the integration of capacitive sensor systems with intelligent regression models can improve the reliability and automate moisture content monitoring in industrial settings. The developed method for measuring moisture content and the system for implementing thereof can be adapted for various types of bulk materials and process environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Improving the accuracy, stability, and reliability of moisture content measurement results for bulk materials: a modern approach based on regression models

  • 期刊论文

  • 发表日期：

    2026年5月6日

查看全文
  • P. I. Kalandarov,
  • B. P. Iskandarov

摘要

This paper discusses ways to improve the accuracy of measuring moisture content in bulk materials, which is critical for ensuring product quality, preservation, and process efficiency in agriculture, food industry, and construction. A brief overview of moisture content measurement techniques is provided, showing that conventional methods, including gravimetric analysis, while highly accurate, lack the ability to provide real-time and continuous monitoring. The paper introduces a newly developed method for measuring moisture content, which utilizes a measurement system that includes the sensor and intellectual components. The sensor component represents a moisture meter comprising capacitive sensors to measure the dielectric constant of the material. The signals from the sensors are transmitted to a data acquisition and pre-processing unit for filtering and normalization. The sensor component provides stable measurements of moisture content for wheat, corn, and sand in the range of 6–25%. The intelligent component of the system includes a regression model, which accounts for the effects of dielectric constant, bulk material density, and ambient temperature on the accuracy of moisture content measurements. It represents a multi-parameter linear regression model implemented using the scikit-learn library (Python). A 10-fold cross-validation was used to assess the stability of the model. Experimental studies showed an average absolute measurement error of less than 1.8% and a determination coefficient of more than 0.89, thus confirming the stability and reproducibility of the system. The presented approach demonstrates that the integration of capacitive sensor systems with intelligent regression models can improve the reliability and automate moisture content monitoring in industrial settings. The developed method for measuring moisture content and the system for implementing thereof can be adapted for various types of bulk materials and process environments.