| As an important branch of smart wear,smart clothing can monitor the wearer's physical condition in real time or protect the human body during special work,and is applied to many fields such as health monitoring,military,etc.With the increasing using,the number of smart clothing is gradually rising.Clothing pressure,temperature and humidity between body and clothing,and the wearer's heart rate are import:ant indicators of clothing comfort evaluation.It is important to develop a measurement system capable of simultaneously collecting these four parameters and a clothing comfort evaluation method based on these four parameters.The system designed in this dissertation uses the battery-powered,low-power chip-MSP430F149 for main controlling,ADS1192 for ECG signal acquiring and conversing and low-power Bluetooth for wireless communication.The LabVIEW host computer collects and displays sensor data in real time.Thirty subjects were selected to measure four parameters under different human actions.The collected EO signal is filtered by the empirical mode decomposition,and the adaptive threshold detection algorithm improves the R wave detection efficiency.Using the method of fuzzy C-means clustering based on improved kernel function to intelligently aggregate the average clothing pressure,extreme pressure difference,extreme temperature difference,extreme humidity difference and change rate of wearer's heart.Using clustering results to train the support vector machine to establish a comfort evaluation model.The research results show that the system has the characteristics of low energy consumption and portability.The fitting linearity of selected film pressure sensor are 0.9989,which reflects that garment pressure has a linear relationship with the output voltage,and the accuracy of temperature,humidity,and heart rate is high.The proposed evaluation method can objectively and accurately evaluate the comfort of smart clothing and guide the production of fit and comfortable smart clothing. |
PDF Full Text Request