Abstract:

Objective With the development and application of flexible sensing technology, there is an urgent need for accurate evaluation of detected results, because there is no mature evaluation standard at present. For the fabric-type sensors, the output of its detection signal is closely related to the changing curvature of the human surface and the physical parameters of the material itself. Therefore, this study attempts to establish a continuous dynamic pressure finite element model for the interface between the human body and clothing, soas to provide theoretical basis for assessing the output accuracy of the fabric-type sensor. It could be used to characterize the accuracy of flexible sensing technology in practical application, so as to further promote the industrialization of smart wearable products.

Method In this research, the human arm is selected as the research object, and a three-dimensional finite element model of the human arm and the elastic pressure arm sleeve is established, based on the finite element principle. The processes of putting the elastic cuff around the human arm and the buckling of the human elbow joint after wearing the elastic arm sleeve are numerically simulated, and the equivalent stress of the soft tissue external surface of the human arm and the elastic pressure arm sleeve changing with time is calculated. Based on the data, a linear regression model was established, and root-mean-square error (RMSE) was selected as the representation index of the linear regression model to characterize the accuracy of the textile flexible sensor test.

Results In order to simplify the model, the fabric and bone were assumed to be isotropic linear elastomer, and the soft tissue was assumed to be isotropic hyper-elastomer. When the pressure arm sleeve was worn to the human arm, 16 points were determined according to the shape of the pressure arm. The effectiveness of the model was verified, and the relative error between the simulated value and the measured value was analyzed. The results showed that except for the outer part of ulna at the lower end of the arm (height 1-t side) and the muscle of the upper arm (height 4-p side), the finite element prediction results of the remaining 15 test points under three different fabrics were basically consistent with the pressure experiment results (0.2% to 8.9%), which can prove the validity of the model. Two types of textile flexible sensors were selected in the market, and the mechano-electric coupling model of the flexible sensors was established, and the electrical signals collected by the sensors were converted into stress values. The dynamic stress curve of the soft tissue surface on the outside of the elbow joint was extracted with time. Linear fitting was carried out according to the feature points, and the fitted curve was used as a linear regression model to characterize the application performance of the collected test data at the elbow joint. The RMSE was selected as the representative index of the regression model, and the accurate performance of the output signal applied to the elbow joint of the two sensors was characterized. The results showed that the regression effect of sensor A was better than that of sensor B. In other words, sensor A is more suitable for the measurement of elbow joint flexion with higher accuracy, indicating that this model could potentially characterize the performance of test accuracy of fabric-type flexible sensors.

Conclusion This paper proposes a method to characterize the accuracy of test data from textile flexible sensor. A finite element method was established to simulate the dynamic pressure of clothing, and a linear regression model was set up to calculate the dynamic pressure using RMSE to characterize the prediction error. For the future research, it is suggested that the establishment of materials and models can be further refined for different parts of the human body to balance the calculation time and get closer to the physiological characteristics of the human body. In order to simulate more complicated working conditions, it is necessary to further understand the biomechanics of the human body in the state of motion. In addition, it is necessary to explore the dynamic pressure model of clothing under multi-physical field coupling to better characterize the performance of fabric-type flexible sensors in practical applications.

Key words: fabric-based sensor, continuous dynamic pressure, numerical simulation, clothing pressure, human arm, elastic pressure arm sleeve