可穿戴健康设备在健康监测领域具有很大应用潜力。其核心部件是传感器，理想的传感元件应该具备高性能、轻薄、柔性和低成本等优点。可穿戴技术与步态分析技术结合，实现了步态在临床医学和运动分析等领域的应用。步态分析系统中鞋垫系统最具竞争优势。目前鞋垫系统存在一些待解决问题。传感器硬度大，贴在鞋垫表面，易打滑磨损且舒适度低；一些鞋垫系统的传感器数目过少，应在足底至少放置15个传感器。基于以上问题，本文研究了轻薄、高性能的柔性压力传感器，并设计柔性智能鞋垫，集成到便携式智能鞋垫系统进行步态分析。为了制备高性能且适合集成到鞋垫系统的压力传感器，以具有优异电学和力学等性能的石墨烯为活性材料，多孔结构SBR材料为基底，通过浸染法得到了石墨烯-SBR基压力传感器。制备过程中对于材料浓度、器件尺寸和基底孔隙密度进行了讨论。通过SEM表征观察到传感器的三维多孔架构和堆叠在骨架的石墨烯薄片。通过Raman表征观察到石墨烯的D、G和2D特征峰。根据足底生物力学结构，在鞋垫基底的16个关键位置嵌入传感器，通过柔性PCB完成电极引线和封装，得到轻薄、柔软的一体化智能鞋垫。对石墨烯压力传感器的关键性能指标进行了测试与分析。测试结果表明，传感器的灵敏度高达1.26?𝑃?−1，测量范围可达300?𝑃?，适用于足底压力测量。传感器具有较好的重复性和一致性，响应时间约120ms，恢复时间约80ms，在20℃-40℃的温度范围内阻值几乎不受温度影响。基于多孔结构对其工作机理进行解释：小压强下石墨烯薄片产生裂缝使阻值增加，中等压强下导电通道增加导致电阻减小，高压强下传感器的结构形变达到极致导致力电性能饱和。分别建立了传感器的简化电阻模型和多孔结构模型，对其力电响应进行分析。应用传感器进行了人体物理信号的监测，如脉搏、手写识别等。采用多路测试仪搭建测试平台，验证智能鞋垫的功能，并进行简单的步态测试。以树莓派系统为处理器，设计16路的AD转换电路，将鞋垫、AD转换电路和树莓派集成得到了智能鞋垫系统，并用Python编程进行数据存储和图形化界面显示。使用系统进行站立测试，结果符合正常人的足底受力分布。通过行走测试，观察到步态周期的四个典型阶段，测试结果也体现了系统较好的重复性。表明了集成智能鞋垫在鞋类设计、临床诊断和运动分析等领域具有广阔的应用前景。
Wearable healthcare devices have great application potential in the field of health monitoring. The core component of the wearable devices is the sensor, and the ideal sensor should have the advantages of high performance, thinness, flexibility, and low cost. The combination of wearable technology and gait analysis technology realizes the application of gait in clinical medicine and sports analysis. The insole system has the most competitive advantages among the gait analysis systems. There are some unsolved problems in the insole system. For example, the sensor has high hardness and is attached to the surface of the insole, which is easy to slip and wear and has low comfort. Some insole systems have too few sensors, and research shows that at least 15 sensors should be placed on the sole. In order to solve the above problems, this paper studied a thin, high-performance flexible pressure sensor, and designed a flexible smart insole, which was integrated into a portable smart insole system for gait analysis.In order to prepare a high-performance pressure sensor suitable for integration into an insole system, a graphene/SBR based pressure sensor is obtained by immersion method using graphene with excellent electrical and mechanical properties as the active material and porous SBR as the substrate material. During the preparation process, the graphene concentration, device size, and substrate pore density were discussed. The three-dimensional porous structure of the sensor and graphene sheets stacked on the framework was observed through SEM characterization. The D, G, and 2D peaks of graphene were observed by Raman characterization. According to the biomechanical structure of the sole, sensors were embedded in 16 key positions of the insole, and the electrode leads and encapsulation were completed through a flexible PCB, and finally, a light, thin and soft integrated intelligent insole was obtained. The key performance indicators of the graphene pressure sensor were tested and the results show that the sensitivity of the sensor is 1.26kPa-1, and the measurement range can reach 300kPa, which is suitable for plantar pressure measurement. The sensor has good repeatability and consistency. The response time is about 120 ms, and the recovery time is about 80 ms. The sensor resistance is almost unaffected at a temperature of 20 ℃ to 40 ℃. The working mechanism is explained based on the porous structure: cracks in the graphene sheet under low pressure increase the resistance value, and the increase of the conductive channel under medium pressure causes the resistance to decrease, and the structural deformation of the sensor reaches the extreme under high pressure, which leads to the saturation of the electromechanical performance. A simplified resistance model and a porous structure model of the sensor were established respectively to analyze the force-electric response. The sensor was used to monitor the physical signals of the human body, such as pulse, handwriting recognition, and so on.A test platform was built through a multi-channel tester, which was used to verify the function of the smart insole and perform a simple gait test. Take the Raspberry Pi system as the processor, and design a 16-channel AD conversion circuit. Integrate the insole, AD conversion circuit, and Raspberry Pi to get a smart insole system. Use Python language programming for data storage and graphical interface display. Using the system to perform a standing test, the results are consistent with the distribution of the plantar force of a normal person. Through the walking test, four typical phases of the gait cycle were observed. The test result also reflects the good repeatability of the system. It shows that the integrated intelligent insole has broad application prospects in the fields of footwear design, clinical diagnosis, and sports analysis.
