The Plantar Momentum Enabled Identification Using Triboelectric Nanogenerator Array

Compared with traditional piezoelectric and piezoresistive sensors,sensors based on Triboelectric Nanogenerator(TENG)have attracted much attention due to its advantages of self-powered and simple structure,but few studies have been done on the output signal acquisition circuit of TENG.In this thesis,single and multiple TENG output signals are successfully detected by self-made circuit.Then a foot recognition system based on TENG sensor array is designed to collect the gait information of human foot during walking,and this system is able to display the relative speed,sequence and contour in real time when foot contact with the ground.By extracting feature parameters and using SVM classification algorithm,the function of character identification can also be realized.The main contents of this thesis are as follows:(1)The working principle of contact-separation mode TENG is analyzed,after that the load voltage output and short-circuit current output of the TENG which is used in this project are tested.The complete derivation process of mathematical model about open-circuit and short-circuit state of TENG is given,also put forward a thought of relationship between momentum and TENG output,and a ball-drop experiment is designed to verify this relationship.(2)After the TENG processing circuit is studied,the output of a single TENG is successfully detected by the negative-feedback integrating circuit,also the multi-channel TENG signals are distinguished by analog switch.The feasibility and practicability of the self-made circuit system are verified by the design of TENG keyboard experiment.(3)The TENG output electrode and the its signal processing circuit are made on one PCB board through combining TENG output electrode with PCB.STM32 MCU built-in ADC unit is used to sample the foot signals collected by the array.The data are sent to PC through serial port.Then PC use a dialog program which is based on MFC to display the foot information dynamically in real time after received data are combined with RGB color scheme.(4)Because of the image result of the foot recognition system is not clear,two improved plans are presented from the aspects of array density and image processing.Firstly,the problem of insufficient sampling points is solved by reducing TENG area and increasing array density.Secondly the image of result is processed by interpolation algorithm.After that,the imaging effect is improved.