Research On Static Comfort Of Automotive Seats Based On Body Pressure Distribution

As an important subsystem of direct contact between automobile and driver,seat comfort is an important factor that must be considered in the process of seat design and manufacture.In recent years,with the improvement of national living standards,higher requirements have been put forward for seat comfort.As an important method for objective evaluation of seat comfort,the relationship between body pressure distribution and subjective comfort is complex and highly non-linear.In this paper,we use intelligent optimization algorithm to explore the quantitative relationship between objective evaluation and subjective evaluation,and build a prediction model of car seat comfort,and combined with finite element human body simulation model to screen the seat foam scheme.The application provides a more simple,accurate and efficient method for the design and evaluation of car seat comfort.Firstly,the subjective and objective evaluation test of static comfort of automobile seat was carried out,and the subjective evaluation table of comfort including all parts of human body was designed.The body pressure distribution data and the subjective evaluation of regional comfort were obtained through subjective and objective evaluation test.The correlation analysis method was used to screen the body pressure indices.Ten body pressure indices were obtained and the normal distribution data were tested.The weights of subjective comfort evaluation items are analyzed by using the method of fuzzy analytic hierarchy process.The weights of 13 subjective comfort evaluation items were obtained,and the overall comfort evaluation results were calculated by weights.Then,the BP neural network(ABC-BP)optimized by artificial bee colony algorithm was used to predict seat comfort.Ten body pressure indices obtained from subjective and objective comfort evaluation tests were used as input and overall comfort evaluation as output.A seat comfort prediction model based on ABC-BP was constructed.89%of the 176 samples were used as training part of the model,and 11%were used as model validation.Compared the predicted results of the real values,the prediction results reach 0.0019 in mean square error and 0.946 in coefficient of determination,the MSE(mean square error)is84.68%lower than that obtained by BP neural network algorithm and the R~2(coefficient of determination)is 42.5%higher than that obtained by BP neural network algorithm.The results show that the prediction model of car seat comfort based on BP neural network optimized by artificial bee colony algorithm is more stable and accurate.Finally,the seat finite element simulation model was established.Through the simulation analysis of the body pressure distribution of three different foam hardness seats,and the body pressure analysis data as input,the comfort comfort score of the three kinds of foam hardness seats was predicted based on the ABC-BP seat comfort prediction model.Practice had proved that the seat finite element simulation model combined with seat comfort prediction model based on ABC-BP can predict the comfort in the early stage of seat design and realize the evaluation of different design schemes.