| With the accelerated aging of China’s population and the rising incidence of chronic diseases such as hypertension,health has become a hot topic in society.The use of wearable electronic devices for health monitoring has become the future development trend and one of the important directions for the development of health industry in various countries.However,the sensing unit of wearable devices needs external power supply,which greatly reduces the comfort and portability of wearable electronic devices.Batteries have a short lifespan and need to be recharged or replaced frequently.More importantly,batteries often contain substances that are harmful to humans,posing a potential threat to human health.Therefore,using external batteries for power supply is not the best solution for powering wearable electronics.Friction nanogenerators have been effectively used in self-driven systems due to their simple structure and various forms,and have received widespread attention.In addition to harvesting energy from the environment to power electronic devices,friction nanogenerators can also be used as selfpowered sensors to respond to external stimuli.Frictional electrical sensors have great medical benefits in the direction of human sensing,but the application of frictional electrical sensors to intelligent health detection systems is still in its infancy.Therefore,this paper focuses on the preparation of frictional electric sensors and their application in the field of human health monitoring,and the main research is as follows:(1)Real-time blood pressure monitoring is essential for the timely diagnosis and treatment of cardiovascular disease.Traditional blood pressure measurement methods predict blood pressure by sensors measuring multiple physiological signals.In this paper,a frictional electric sensor based on a dual sandwich structure and a new blood pressure prediction method are proposed,and a continuous,cuff-less blood pressure monitoring system is also developed.The designed sensor with dual sandwich structure achieves a high sensitivity of 0.89 V/k Pa over a linear range of 0 to 35 k Pa,which is more than twice that of the conventional sensor with single electrode structure.The response time of the sensor is only 32 ms,so that a weak pulse signal can be easily captured.We fuse user background information and pulse wave signals and combine them with deep learning techniques for blood pressure prediction.To improve the prediction accuracy,we propose a deep learning model with a multi-network structure.The mean absolute error and standard deviation of error of this model for systolic and diastolic blood pressure are as low as 3.79 ± 5.27 mm Hg and 3.86 ± 5.18 mm Hg.(2)Wearable electronic devices based on friction nanogenerators have become more powerful with the rapid development of flexible frictional electrical pressure sensors.However,most triboelectric sensors can only measure in a small pressure range,which limits their application in multiple scenarios.Here,we propose a wide-range flexible triboelectric pressure sensor through the difference of material Young’s modulus and the optimization of surface structure angle.By analyzing the effect of the structural angle of the material surface on the sensor performance,a triboelectric sensor based on the optimal combination of inclination angles is obtained.The sensitivity of the sensor with the optimal angle combination is 249.32 m V/k Pa in the low pressure range(0-75 k Pa)and36.24 m V/k Pa in the high pressure range(75-450 k Pa).The proposed wide range,low detection limit(200 mg)and fast response(26 ms)of the triboelectric sensor can be applied not only for weak physiological signal detection but also for larger pressure transmission.It can also be applied to larger pressure sensing scenarios.We developed a human action recognition system based on plantar pressure sensing and proposed a convolutional gated cyclic unit model to recognize four human actions with high accuracy(99.42%). |
PDF Full Text Request