| Breathing signals are one of the important physiological signals of the human body,and its effective detection can help people understand their physical health and prevent other diseases such as respiratory diseases and other diseases caused by breathing in time.Traditional contacttype breath signal detection methods such as polysomnograph detection method and piezoelectric detection method of piezoelectric ceramics often require long-term wear of sensors and the use of rigid materials for their materials,which brings great discomfort to the patient’s physiology and psychology,and has certain limitations for human breathing signal detection.Therefore,effective respiratory signal detection methods and treatment analysis methods have important research significance for the diagnosis of respiratory diseases.This thesis mainly studies the production of flexible piezoelectric sensors and the acquisition and processing methods of breathing signals,which are summarized as follows:Firstly,in view of the disadvantages of rigid material sensors,this thesis proposes a design scheme for a composite flexible respiratory piezoelectric sensor based on polyvinylidene fluoridetrifluoroethylene(PVDF-TrFE)/zinc oxide(ZnO).Using the electrospinning device,a PVDFTrFE/ZnO composite flexible piezoelectric film containing ZnO of 0wt%(pure PVDF-TrFE),0.1wt% and 0.2wt% was prepared,and the nanoscale micromorphology structure of different film sample fibers was observed under scanning electron microscopy,and the influence of ZnO with different mass fractions on the microstructure of composite piezoelectric film fibers was explored.Secondly,the piezoelectric film is designed and packaged into a flexible piezoelectric sensor with the piezoelectric film as the functional layer.The open output voltage of the composite piezoelectric sensor with different mass fractions of ZnO was tested by the electrical response output experiment,and the comparison showed that the piezoelectric sensor containing 0.2wt% of ZnO had the best piezoelectric performance,and the open output voltage reached 19.5V,which was 51.2% higher than that of the pure PVDF-TrFE piezoelectric sensor.For the optimum piezoelectric sensor,response time test,stability test and elbow motion signal detection were performed.Using this piezoelectric sensor as the front-end collector,a signal acquisition system was built to realize the acquisition of human breathing signals.Third,the signal acquisition system is used to collect breath signals,and the dynamic detection effect of the piezoelectric sensor for several breath signals is verified.Aiming at the noise problem contained in the original breath signal,based on the principle of wavelet transform,this thesis proposes a denoising algorithm for the bivariate threshold function,which takes the two parameters of mean squared error and signal-to-noise ratio after denoising as the performance indicators,indicating that the denoising effect of the algorithm is better than that of the traditional soft threshold and hard threshold functions.Finally,a combination of first-order differential algorithm and threshold setting method is used to mark the crests and troughs of the breath signal after denoising,and extracts the two time domain information of the crest trend and the breath wave period according to the marker.Aiming at apnea signal detection,a method for identifying apnea based on an equivalent model is proposed,and the accuracy of the method is verified by 10 sets of experiments. |
PDF Full Text Request