【作者】 王军；

【导师】 刘富；

【作者基本信息】 吉林大学 ， 电子信息（专业学位）， 2024， 硕士

【摘要】 我国提出的《中国制造2025》、《“十三五”国家科技创新规划》等文件,都将“传感器技术”列为重要的发展方向。其中扭矩传感器在工业生产中发挥着重要作用,它可以用于监测机械装配过程中的扭矩、检测发动机输出扭矩以及检测风力发电机组叶片安装扭矩的准确性等等。柔性传感器相较于传统刚性传感器具有诸多优势,包括高度的适应性、可穿戴性、轻量化、柔软性等特点。这使得柔性传感器在人机交互、医疗健康监测、智能可穿戴设备等领域展现出广泛的应用前景。将扭矩传感器与柔性传感设计相结合可以更好地为工业生产提供高性能的扭矩检测方案。在当前社会对智能化、便携式设备需求的推动下,柔性传感器的发展变得愈加迫切,为实现更加智能、便捷、舒适的生活方式提供了重要支持。近年来,从生物独特的感知能力中寻求柔性传感器的设计灵感成为了热门研究方向。如典型生物蝎子,其腿部的缝结构微振动感受器使其具有远超其他生物的微振动信号感知能力。通过模仿蝎子缝单元感知原理研发的柔性传感器具备极高的检测灵敏度。但在现有研究中,仿生缝结构的设计与应用大都局限于简单的应变检测,缺乏对扭矩等特殊应变的适应性设计,同时柔性传感器对复杂检测面的适应性也存在很大的局限性,因此,本文以蝎子缝感受器的微观结构为仿生模本,研究缝结构柔性传感器在大区率表面扭矩检测场景下的应用,通过有限元的仿真实验、数学模型推导论证了其可行性,通过合理的工艺设计制备了仿生缝结构的蛇形扭矩传感器并对其开展了相关的性能测试及应用实验。在结构设计上,结合缝感知结构与蛇形延展性结构的蛇形缝结构能较好的适应复杂的检测面与高精度的检测需求。在此基础上,本文选取了与传感器结构相关的关键参量进行仿真实验设计,控制单一变量进行有限元仿真实验,从有限元的角度探究了传感器结构参数与性能的关系。在此基础上本文还建立了仿生蛇形缝单元临界闭合状态响应的数学模型,为揭示本文传感器的检测机理提供了理论基础。设计工作之后本文进行了进一步的传感器工艺设计,对柔性材料加工、缝单元刻蚀、金属层生长、封装集成等工艺进行了进一步的研究,提出了仿生蛇形缝结构柔性扭矩传感器的制造方法,并对传感器性能进行了包括灵敏性、耐久性、梯度检测特性等的多种测试,验证了仿生蛇形缝单元临界闭合状态响应模型的正确性及传感器制备工艺的合理性。最后,为了验证该传感器在实际应用中的价值,本文设计进行了两个应用实验,第一个是低扭矩范围内的汽车驱动轴扭矩检测实验,把本文传感器与商用的电阻应变片在相同的实验条件下进行低扭矩范围的性能测试,将实验结果在同一坐标系下进行对比,证明了在该实验条件下本文传感器的检测性能更好,由此也验证了本文传感器在高精度扭矩检测领域的应用价值。第二个应用实验为人体手腕翻转模态检测实验,将传感器阵列固定于手腕尺骨上部,实时采集传感器的阻值变化情况,利用神经网络(Neural Network,NN)和支持向量机(Support Vector Machines,SVM)同时进行模型训练,选取识别准确率高者作为分类识别算法进行实时模态辨识,结合构建的实时显示界面完成了人体手腕模态的检测实验,验证了本文传感器的应用价值。综上所述,本文根据蝎子缝感受器的微振动感知原理,设计了一种可以适应复杂检测面、可进行高精度检测的仿生传感器,通过有限元仿真实验的设计和仿生蛇形缝单元临界闭合状态响应模型的构建,从理论上证明了传感器设计的合理性。通过合理的工艺设计完成了仿生蛇形缝结构的柔性扭矩传感器的制备。同时对传感器进行了性能指标测试及应用实验测试,验证了模型及工艺的正确性,丰富了仿生传感器设计与应用方面的研究。更多还原

【Abstract】 Documents such as "Made in China 2025" and "National Science and Technology Innovation Plan for the 13 th Five-Year Plan" proposed by China have identified "sensor technology" as an important direction of development.Among them,torque sensors play a significant role in industrial production,as they can be used to monitor torque during mechanical assembly processes,detect engine output torque,and ensure the accuracy of torque during wind turbine blade installation,among other applications.Compared to traditional rigid sensors,flexible sensors have many advantages,including high adaptability,wearability,lightweight,and flexibility.These characteristics make flexible sensors promising in various fields such as human-machine interaction,medical health monitoring,and smart wearable devices.Integrating torque sensors with flexible sensor designs can provide better torque detection solutions for industrial production.With the increasing demand for intelligent and portable devices in society,the development of flexible sensors has become more urgent,providing essential support for achieving a more intelligent,convenient,and comfortable lifestyle.In recent years,seeking inspiration for flexible sensor design from the unique sensing capabilities of biological organisms has become a hot research direction.For instance,typical biological organisms like scorpions possess highly sensitive microvibration receptors in their leg joints,far exceeding the micro-vibration signal perception capabilities of other organisms.Flexible sensors developed by mimicking the sensing principle of scorpion seam units exhibit extremely high detection sensitivity.However,in existing research,the design and application of biomimetic seam structures are mostly limited to simple strain detection,lacking adaptive designs for special strains such as torque.Additionally,flexible sensors have significant limitations in adapting to complex detection surfaces.Therefore,this paper takes the microstructure of scorpion seam sensors as a biomimetic model to study the application of seam structure flexible sensors in torque detection scenarios on large-area surfaces.The feasibility is demonstrated through finite element simulation experiments and mathematical model derivation.A biomimetic serpentine torque sensor with seam structures was fabricated through rational process design,and relevant performance tests and application experiments were conducted.In terms of structural design,the serpentine seam structure,combining seam sensing structures with serpentine extensibility structures,can better adapt to complex detection surfaces and high-precision detection requirements.Based on this,this paper selected key parameters related to sensor structure for simulation experiment design,conducted finite element simulation experiments by controlling single variables,and explored the relationship between sensor structure participation and performance from the perspective of finite elements.Furthermore,this paper established a mathematical model for the critical closure response of biomimetic serpentine seam units,providing a theoretical basis for revealing the detection mechanism of the sensor.After the design work,this paper further conducted sensor process design,including research on flexible material processing,seam unit etching,metal layer growth,encapsulation integration,etc.The manufacturing method of the biomimetic serpentine seam structure flexible torque sensor was proposed,and various tests including sensitivity,durability,and gradient detection characteristics were conducted to verify the correctness of the biomimetic serpentine seam unit’s critical closure response model and the rationality of sensor preparation technology.Finally,to validate the value of the sensor in practical applications,two application experiments were designed and conducted.The first experiment was a torque detection experiment on automobile drive shafts within a low torque range.The performance of this paper’s sensor was tested under the same experimental conditions as commercial strain gauges within the low torque range,and the experimental results were compared on the same coordinate system,proving better detection performance of this paper’s sensor under these experimental conditions.This also verified the application value of this paper’s sensor in the field of high-precision torque detection.The second application experiment was a detection experiment on human wrist rotation modes.The sensor array was fixed on the upper part of the wrist bone,and real-time changes in sensor resistance were collected.Neural network(NN)and support vector machine(SVM)models were trained simultaneously,and the one with higher recognition accuracy was selected as the classification recognition algorithm for real-time mode identification.Combined with the constructed real-time display interface,the experiment on human wrist mode detection was completed,verifying the application value of this paper’s sensor.In summary,based on the micro-vibration sensing principle of scorpion seam sensors,this paper has designed a biomimetic sensor that can adapt to complex detection surfaces and perform high-precision detection.The rationality of sensor design has been theoretically proven through finite element simulation experiment design and the construction of a biomimetic serpentine seam unit’s critical closure response model.The fabrication of the biomimetic serpentine seam structure flexible torque sensor was completed through rational process design.Performance index tests and application experiments were conducted on the sensor,verifying the correctness of the model and technology,enriching the research on biomimetic sensor design and application.更多还原