Impact And Evaluation Of Traffic Safety In Superlong Highway Tunnels Based On Warning Sound Factors

To implement the strategy of building a strong transportation country and improve traffic safety inside tunnels,most existing research focuses on improving the driver’s visual environment,conducting extensive research and analysis on tunnel lighting,control methods,interior materials,and transportation facilities.Hearing as the second main way for drivers to obtain information,and there are few studies based on it.In recent years,some tunnel operators have broadcasted fire sirens continuously inside tunnels in hopes of improving driver fatigue or distractions through sound wave impact.However,there is a lack of in-depth scientific research on the selection of warning sounds,the mechanism of sound wave impact on driver’s physiological characteristics and driving Behavior,and whether it helps improve traffic safety.This paper takes warning sounds as the research object.Firstly,the warning sounds to be studied are selected through questionnaire surveys.Then,a combination of indoor,field experiments and theoretical analysis are used to systematically analyze the impact of warning sounds on drivers’ physiological characteristics,driving behavior,and driving speed.Finally,a weight system was established through principal component analysis and entropy method,and the fuzzy comprehensive evaluation method was used to evaluate the traffic safety inside the tunnel under the action of warning sounds.The main research work and conclusions of this article are as follows:(1)Based on the coherent model and ISO sound wave model,a theoretical model of sound wave propagation in tunnel driving scenes under the act of warning sound is derived and established.Through the services of professional questionnaire survey institutions,the top five warning sound types to be studied are selected:artificial sound"Please pay attention to safety",police car alarm,ordinary car horn,soothing piano music(such as canon),and fire truck alarm.The factors that are significantly correlated with the acceptance and effectiveness prediction results of warning sounds are calculated by Spearman correlation coefficient.(2)Through field experiments,the impact of warning sounds on drivers’ physiology,driving behavior,and driving speed was studied.The impact of three warning sounds(fire alarm,dynamic music,and voice command)on drivers’ pupil diameter and blink time was analyzed,as well as the paired t significance test.Based on Markov chains,the characteristics of drivers’ gaze point transfer caused by warning sounds were further analyzed.Fourier transform and wavelet denoising were used to process warning sound signals,and the possibility of their impact on drivers’ visual physiology was deeply explored from the perspectives of sound waveforms and frequencies.In addition,the impact of fire alarm sounds on driving speed,accelerator pedal opening,and vehicle acceleration was analyzed,and paired t-tests were conducted to test the significance.The relationship between warning sound intensity and tunnel length for superlong highway tunnels was fitted and established,and Pearson coefficients were used to further analyze the correlation between driving speed,accelerator pedal opening,and vehicle acceleration with warning sound intensity,as well as the sensitivity to warning sound intensity,comparison shows that acceleration is the most sensitive to the loudness of warning sounds.(3)Using an orthogonal design indoor experiment,systematic physiological experiments were conducted to investigate the effects of warning sound parameters such as sound type,loudness,playing duration,and playback rate on EEG(β、(θ+α)/β)and heart rate variability(RMSSD、SDSD、pNN50).Based on Fourier transform and wavelet denoising,the processing of EEG and ECG signals was conducted to analyze the variation patterns of evaluation indicators and paired t significance tests.Furthermore,range analysis and factor level trend analysis were used to further analyze the primary and secondary orders of the impact of various parameters and levels of warning sounds on EEG and heart rate variability.The study found that the order of the impact of warning sound on vigilance level is loudness,sound type,playback rate,and playback duration.The order of the impact on fatigue level is playback duration,loudness,sound type,and playback rate.(4)Considering both subjective and objective factors,a traffic safety weight system under the influence of warning sounds was constructed using principal component analysis and entropy method.Based on the measured driving speed data of multiple sections inside the tunnel,a speed model was established for the geometric parameters of tunnel alignment.With it,the operating speed coordination evaluation was carried out.Combined with existing research results,the other evaluation indicators included in the weighting system were evaluated at three levels.Finally,the fuzzy comprehensive evaluation method was used to comparatively evaluate tunnel traffic safety with or without warning sounds.The evaluation shows that the warning sound has a certain improvement effect on the traffic safety of superlong highway tunnels,but its improvement effect is not significant enough within the safety level range determined in this article.