基于蛛网共振感知原理的可穿戴柔性应变传感器与棉秆起拔力原位检测

Wearable Flexible Strain Sensor Based on Principle of Spider-web Resonance Sensing and In-situ Detection of Cotton Stalk Pulling Force

摘要:

针对棉秆拔除过程起拔力无法精准获取问题，本文基于蛛网共振感知原理，设计一种新型棉花茎秆可穿戴仿生柔性应变传感器，对棉秆起拔力进行原位检测。使用Solidworks软件设计2种蛛网仿生结构，并通过有限元仿真确定最佳蛛网结构几何参数（半径梯度为36、38、40 mm，线宽和间距均为1 mm），以热塑性聚氨酯（TPU）为弹性基底，采用丝网印刷技术制备并封装柔性传感器，搭建数据采集系统对传感器的灵敏度、响应时间、漂移和稳定性等性能进行测试。测试结果表明，传感器灵敏度系数（GF）随应变呈线性增长，响应时间为250 ms，恢复时间为590 ms，经过100次拉伸循环，电阻变化表现出良好的稳定性，应变在0%~12%范围内最大迟滞误差为3.6%，且温度波动-15~25℃时电信号变化微弱。通过传感器-棉秆静态载荷标定试验，将棉花茎秆的力信号转换成电信号，两者输出信号变化趋势一致，并建立了起拔力-电阻变化模型（R2=0.998），验证试验结果平均相对误差不大于4.78%。田间测试结果验证了传感器在棉秆拔除过程能实时监测，准确感知起拔力动态变化。研究结果可为棉秆收获装备优化改进提供数据支撑，为柔性传感技术在各种作物原位检测的应用提供参考。

Abstract:

In response to the problem that the pulling force during cotton stalk removal cannot be accurately obtained, based on the principle of spider-web resonance perception, a type of wearable bionic flexible strain sensor for cotton stalks was developed to conduct in-situ detection of the pulling force. Two types of web bionic structures were designed using Solidworks software, and the optimal geometric parameters of the web structure (radius gradient of 36 mm, 38 mm, and 40 mm, line width and spacing of 1 mm) were determined through finite element simulation. Using thermoplastic polyurethane (TPU) as the elastic substrate, the flexible sensor was fabricated and encapsulated by screen printing technology, and a data acquisition system was set up to test the sensor's sensitivity, response time, drift, and stability. The test results showed that the sensor's sensitivity coefficient (GF) was increased linearly with strain, the response time was 250 ms, the recovery time was 590 ms, and after 100 stretching cycles, the resistance change exhibited good stability, with a maximum hysteresis error of 3.6% within a strain range of 0% to 12%, and the electrical signal change was weak when the temperature fluctuated between -15℃ and 25℃. Through the static load calibration test of the sensor and cotton stalks, the force signal of the cotton stalks was converted into an electrical signal, and the output signal change trends of both were consistent. A tensile force-resistance change model (R2=0.998) was established, and the average relative error of the verification test results was no more than 4.78%. The field test results verified that the sensor could accurately monitor the dynamic changes of the pulling force during the cotton stalk removal process in real time. The research results can provide data support for the optimization and improvement of cotton stalk harvesting equipment and offer a reference for the application of flexible sensing technology in-situ detection of various crops.