Design Of Flexible Piezoelectric Thin Film Sensor And Its Application In Sleep Monitoring

Sleep is an essential physiological activity for the human body,and good sleep can make people energized.Recognizing the sleep state is the foundation of sleep research.The traditional method is to use a multi-channel sleep monitor to stage sleep.However,various rigid sensors can cause psychological burden,and the complex operation process and expensive cost make it unsuitable for daily life.Therefore,researching a simple and convenient sleep monitoring device has become a current focus of attention.Ballistocardiogram(BCG)signals contain a wealth of information about human physiological activities and can be used to identify sleep stages.Therefore,the detection and analysis of BCG signals have important significance.Firstly,in terms of the acquisition method of BCG signals,this article proposes a scheme for a flexible piezoelectric sensor based on a high-voltage electrospinning process to prepare poly(vinylidene fluoride-trifluoroethylene)/lead zirconate titanate(P(VDF-Tr FE)/PZT)composite films.The authors designed a high-voltage electrospinning platform with a temperature and humidity controller,and prepared P(VDF-Tr FE)/PZT piezoelectric composite films with different PZT mass fractions based on this platform.Different samples were characterized by scanning electron microscopy(SEM)and X-ray diffraction(XRD),and the results showed that the β-phase of P(VDFTr FE)and the perovskite phase of PZT could be correctly formed.With the addition of PZT,the nanofibers became thinner and more evenly distributed.In addition,a self-built vibration testing platform was used to perform piezoelectric output tests on samples with different PZT mass fractions,and the results showed that the P(VDF-Tr FE)/PZT composite piezoelectric film containing 30% PZT had the largest open-circuit voltage and short-circuit current.Compared with the pure P(VDF-Tr FE)piezoelectric film,its output increased by 164% and 196%,respectively.Secondly,the piezoelectric film was encapsulated as the core layer to form a flexible piezoelectric sensor.Sensitivity tests were performed on flexible piezoelectric sensors containing different PZT mass fractions using a self-designed sensitivity testing platform.The results showed that the addition of PZT nanoparticles was beneficial to improve the piezoelectric performance of the sensor,and the sensitivity of the flexible piezoelectric sensor containing 30% PZT was increased by150% compared to the pure P(VDF-Tr FE)piezoelectric sensor.The authors designed a BCG signal acquisition system for the 30% PZT P(VDF-Tr FE)/PZT flexible piezoelectric sensor,which mainly includes a charge amplifier,a data acquisition card,and a Lab View virtual platform.Based on this platform,the acquisition,filtering,and storage of BCG signals were realized.Then,aiming at the problem of noise in the BCG signal,the author used the semi-soft threshold wavelet function to denoise the BCG signal.The results showed that the semi-soft threshold function overcame the disadvantages of the soft and hard thresholds,and obtained clearer and more accurate BCG signals.Finally,a 4-stage model of sleep was studied based on support vector machines.The J-peaks in the denoised BCG signals were detected and marked by differential threshold method,and 18 features of heart rate variability(HRV)in time domain,frequency domain,and non-linear domain were extracted.The authors used a grid search method based on support vector machines to optimize the parameters and constructed a sleep four-classification model using the best parameter combination.A total of 2000 labeled samples from the MIT-BIH database were used for training,and 200 samples were used for testing.The results showed that the average accuracy of the sleep four-classification model reached 72.5%.The study shows that the P(VDF-Tr FE)/PZT flexible piezoelectric sensor prepared in this paper can acquire human BCG signals,and after filtering,denoising,and feature extraction,it can be used for sleep staging,which has certain feasibility and advancement.It has potential applications in the field of portable sleep monitoring.