ANALYTICAL CHEMISTRY
·
2023
·
IF 7.4
Yanzuo Jiang,Di Huang,Hongjia Zhang,Ting Jiang,Wei Xu
Similar to smartphones, smart or automatic level is also a critical feature for a miniature mass spectrometer. Compared to large-scale instruments, miniature mass spectrometers often have a lower mass resolution and larger mass drift, making it challenging to identify molecules with close mass–charge ratios. In this work, a miniature mass spectrometer (the Brick-V model) was combined with intelligent algorithms to realize rapid and accurate identification. This Brick-V mass spectrometer developed in our lab was equipped with a vacuum ultraviolet photoionization (VUV-PI) source, which ionizes volatile organic compounds (VOCs) with minor fragments. Machine learning would be especially helpful when analyzing samples with multiple characteristic peaks. Four machine learning algorithms were tested and compared in terms of precision, recall, balanced F score (F1 score), and accuracy. After optimization, the multilayer perceptron (MLP) method was selected and first applied for the automatic identification and differentiation of ten different fruits. By recognizing the pattern of multiple VOCs diffused from fruits, an average accuracy of 97% was achieved. This system was further applied to determine the freshness of strawberries, and strawberry picking at different times (especially during the first 24 h at room temperature of winter) could be well discriminated. After building a database of 63 VOCs, a rapid method to identify compounds in the database was established. In this method, molecular ions, fragment ions, and dimer ions in the full mass spectrum were all utilized in the machine learning program. A satisfactory prediction accuracy for the 63 VOCs could be achieved (>99%).
研究领域:
分析化学
·
人工智能
✓ 已验证
DOI
PubMed
高影响因子
Advanced Composites and Hybrid Materials
·
2024
·Vol. 8
·期 1
·1-14
·
IF 23.2
Zhang Minyue,Liu Si,Liu Shun,Jia Gaoen,Zhan Pengfei,Liu Chuntai,Shen Changyu,Liu Hu
In recent years, as wearable electronics continue to advance toward flexible, lightweight, and versatile designs, flexible pressure sensors with wide response ranges and high sensitivity have shown tremendous research value and application potential. In this study, we fabricated TPU-based flexible pressure sensors with a multistage gradient porous structure using layer-by-layer freezing and solvent templating techniques. Due to the layered differences in Young’s modulus from varying porosities, these sensors exhibit high pressure sensitivity ( S , S MAX = 34.08 MPa −1 ) and can accurately distinguish stresses across a wide range (0–1.2 MPa). Additionally, they demonstrate rapid response and recovery times (140 ms), durability over 3000 compression cycles, and the ability to detect both subtle movements (facial expressions and swallowing) and larger actions (joint bends, walking, and running). Furthermore, we developed a smart glove using these gradient-structured pressure sensors combined with a K-nearest neighbor (KNN) algorithm, enabling accurate identification of various fruit types. Notably, the TPU sensors also exhibit excellent thermal insulation and Joule heating properties, making them effective for human thermal management even in extreme temperatures.
研究领域:
材料科学
·
电子工程
·
人工智能
✓ 已验证
DOI
高影响因子
CHEMICAL ENGINEERING JOURNAL
·
2025
·
IF 13.2
Zhejun Jin,Shuheng Dong,Xueqian Liu,Xuqi Zheng,Jingjing Guo,Ho-Kun Sung,Leonid Chernogor,Zhao Yao,Yuanyue Li,Yang Li
The development of highly sensitive and versatile flexible pressure sensors presents significant potential for advancing precision in human-computer interaction systems, particularly in addressing challenges associated with discriminating objects possessing nearly identical visual characteristics. In this work, we fabricated a hierarchical microstructure-enhanced flexible capacitive pressure sensor through synergistic integration of polyvinylpyrrolidone (PVP), polydimethylsiloxane (PDMS), and laser-induced graphene (LIG). A sandpaper-assisted electrospinning technique was employed to create the dielectric layer consisting of PDMS@PVP nanofiber membrane (NM) with irregular concave microstructures, while porous LIG electrodes were precisely patterned onto PVP NMs using screen-printing technology. The incorporation of PDMS effectively enhanced interfacial adhesion and structural plasticity of PVP NM, while the sandpaper-templated concave microstructures in the NM dielectric layer substantially improved sensor sensitivity, particularly in medium-high pressure regimes. The optimized sensor demonstrates exceptional performance metrics including: remarkable sensitivity (0.338 kPa −1 in 0–1 kPa range), broad detection range (0–200 kPa), ultrafast response/recovery times (31/41 ms), ultralow detection limit (2.9 Pa), and outstanding durability (>3000 cycles). Capitalizing on these advancements, the sensor was successfully integrated into a robotic manipulator system, and a fruit identification system was developed, incorporating an embedded system with artificial intelligence algorithms. This integrated system achieves dual discrimination capabilities: distinguishing fruits with nearly identical size/weight characteristics (95.83 % accuracy) and identifying ripeness stages of individual fruits (100 % accuracy). This study establishes a robust technological strategy to overcome tactile recognition limitations for visually indistinguishable objects, demonstrating transformative potential for intelligent human-machine interface applications.
研究领域:
传感器与传感技术
·
材料科学
·
纳米技术
·
机器人学
·
人工智能
·
人机交互
✓ 已验证
产品关联
DOI
高影响因子
危险品化学品经营许可证（带存储）