Scholarlink

<p>The study investigates the design, fabrication, and testing of an advanced microfluidic system for precision soil nutrition detection. It presents an interpretable AI (IAI) as a hybrid approach combining MHBACN_LIME-MSHO to optimize microfluidic systems for enhanced soil nutrient detection. RSM is used to model the relationship between the factors such, as Reynolds number, reagent concentration, applied pressure, and velocity, and the responses include flow rate, mixing index, effective mixing length, and nutrient density. The analysis of RSM models shows high explanatory power with strong R² values exceeding 0.9, indicating a reliable predictive model. A desirability value of 0.972 indicates that the optimal solution. The IAI demonstrates superior performance when compared to traditional methods such as DNN-BA and DT-GWO. The proposed microfluidic chip was validated using real soil samples, demonstrating accurate NPK detection under practical conditions. The results from error analysis, including MSE (0.007), highlight the robustness and precision of the proposed model. The proposed IAI framework not only optimizes system parameters but also provides interpretability, making it a valuable tool for real-world applications. This work offers a comprehensive, reliable, and optimized solution for microfluidic systems.</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

Design, fabrication and testing of advanced microfluidic systems for precision soil nutrition detection based on interpretable AI

  • 期刊论文

  • 发表日期：

    2025年10月17日

查看全文
  • Sachin M. Khomane,
  • Pradeep V. Jadhav,
  • Seema S. Khomane

摘要

The study investigates the design, fabrication, and testing of an advanced microfluidic system for precision soil nutrition detection. It presents an interpretable AI (IAI) as a hybrid approach combining MHBACN_LIME-MSHO to optimize microfluidic systems for enhanced soil nutrient detection. RSM is used to model the relationship between the factors such, as Reynolds number, reagent concentration, applied pressure, and velocity, and the responses include flow rate, mixing index, effective mixing length, and nutrient density. The analysis of RSM models shows high explanatory power with strong R² values exceeding 0.9, indicating a reliable predictive model. A desirability value of 0.972 indicates that the optimal solution. The IAI demonstrates superior performance when compared to traditional methods such as DNN-BA and DT-GWO. The proposed microfluidic chip was validated using real soil samples, demonstrating accurate NPK detection under practical conditions. The results from error analysis, including MSE (0.007), highlight the robustness and precision of the proposed model. The proposed IAI framework not only optimizes system parameters but also provides interpretability, making it a valuable tool for real-world applications. This work offers a comprehensive, reliable, and optimized solution for microfluidic systems.