Research On The Warning Sound Of Electric Vehicles In All Working Conditions Based On Sound Quality

With increasingly stringent energy conservation and emission reduction regulations as well as motivating national policies appearing on the stage,future smart transport in urban area will be electric vehicles(EV).Due to the driving characteristics of electric vehicles,pedestrians outside the vehicle cannot realize the passing of the vehicle being drove at a low speed,which may cause safety hazards.In order to improve safety,governments of various countries have issued regulations to install pedestrian warning sound systems to improve the detectability of electric vehicles and avoid hidden dangers.Based on this background,in this article we researched the acoustic characteristics of electric vehicles combining with the sound quality evaluation,developed and studied the warning sound of electric vehicles under low-speed driving conditions.Drivers and pedestrians outside the car can judge the operating state of the car according to the characteristics of the warning sound signal,so that pedestrians outside the car can effectively avoid traffic risks.The main research content includes:First of all,according to the trend of electric vehicle warning sound development and the main issues that consumers are concerned about,the electric vehicle warning sound intention questionnaire was designed.By analyzing the result,high-speed sound of the electric vehicle was determined as the structural sound source of the low-speed warning sound.The composition of the sound source in the electric vehicle as well as the mechanism of its production was analyzed.Finally,a sound source collection experiment was performed in a semi-anechoic room,and sound signals of various working conditions were obtained to be the reserved sound sources.Secondly,this article introduced various definitions of sound quality psychoacoustic objective parameters and their calculation models as well as subjective evaluation method.Through comparative analysis,the grade scoring method was determined as the subjective evaluation method of this article.The GA-BP model was selected as the sound quality prediction model of this article.Finally,the weight analysis of psychoacoustic objective parameters was carried out,and parameters that have a great impact on the irritability were found to be re-modeled so that the efficiency of the subsequent warning sound synthesis would be improved.Thirdly,a band-stop digital filter was designed to determine the sensitive frequency band of the sound quality in 24 Bark,A band-pass digital filter was designed to divide the sound source signal into 24 signals with 24 Bark as the limit.A corresponding gain coefficient was introduced to each Bark according to the determined sound quality sensitive frequency band and working condition characteristics.The gain coefficient would be adjusted to change the amplitude after further considering the frequency band that affects the environment and the hearing difference of the elderly and normal people.To determine the optimal gain coefficient vector and synthesize new sound source signal,the gain coefficient was adjusted after comprehensive consideration to change the amplitude.Psychoacoustic parameters of the synthesized sound source were calculated and brought to the GA-BP sound quality prediction model to get the subjective irritability,and the best sound was chosen as the warning sound.Finally,the electric vehicle warning sound system was designed.The software program design and hardware circuit design based on the TMS320F2812 chip were completed on the CSS,and the warning sound designed in this paper was verified through the actual vehicle test.According to the actual vehicle test results,the warning sound designed in this paper meets the requirements of the national standard,with good sound quality and satisfying warning effects under all working conditions.The loudness of the warning sound under each working condition increases with the increase of the vehicle’s speed,and the frequency corresponding to the max amplitude of each working condition also changes with the change of the working condition.