Study On The Correlation Between Subjective And Objective Evaluations For Ride Comfort Of Passenger Vehicles

As an extension of living space,passenger cars have become the main mode of transport for the time being,and people have gradually improved their requirements for ride comfort.Ride comfort of passenger cars can be evaluated by subjective evaluation,objective evaluation and computer simulation.The relationship between subjective evaluation and objective evaluation of ride comfort of passenger vehicles is mainly studied,which aims to establish a correlation model of subjective evaluation indicators on objective measured parameters,achieve subjective evaluation objectively,and eliminate the disadvantages of subjective evaluation uncertainty.By analyzing the domestic and international riding comfort evaluation theory,anew subjective evaluation system and objective evaluation method for riding comfort is established.Based on CAE J1060 and introducing 0.25-point interval scoring criteria,thirteen representative subjective evaluation indicators of ride comfort are used.And vibration and noise on riding comfort are fully considered.Entropy method is used to determine the weights of subjective evaluation indicators,which make the overall subjective evaluation effective.The results show that this subjective evaluation system is more biased towards vibration comfort.The random input driving test and pulse input driving test are determined as the objective evaluation methods for the passenger comfort.At the same time,the passenger car interior noise measurement test is conducted.And objective evaluation indexes of the ride comfort of the passenger vehicle are established.Taking 7 test vehicles as the research object,subjective and objective evaluation tests of ride comfort are carried out and subjective index scores and objective measured parameters are obtained.Through regression analysis and BP neural network,the relationship between subjective evaluation and objective evaluation is determined,and the correlation model of subjective indicators on objective measured parameters is established.After verifying and comparing the regression model and the BP neural network model,the results show that both correlation models have good ability of fitting and forecasting.The regression analysis model accuracy is slightly higher than the BP neural network model.It is concluded that the regression analysis model is more suitable for solving the problem of small sample size of subjective evaluation and objective evaluation of passenger car ride comfort.Regression analysis shows that among the 13 subjective evaluation indicators,there are two indicators with the corresponding objective measured parameters showing a binary linear negative correlation,these two indicators are body pitching and rolling.There are eight indicators showing a one-way linear negative correlation with the corresponding objective measured parameters,these indicators are impact sensation,jerks,vibration absorption ability,edge sensitivity,body damping characteristics,roar,noise roughness,and rolling noise.Among them,rolling noise is linearly negatively correlated with Average sound level at the right ear of the driver at a uniform speed of 60 km/h.There are three indicators with non-linear relationship with the corresponding objective measured parameters,these indicators are ups and downs,surges,nods/posts.It can be concluded that the maximum(absolute value)acceleration response value of the pulse input to the driver's seat backrest is a key factor affecting the subjective evaluation.The axially weighted acceleration of the axis at the driver's seat is also a major factor affecting the subjective evaluation.