Research On Design Method Of Electric Vehicle Warning Sound Quality

With the development of the national economy,the production and sales of the automobile industry has been steadily and rapidly improved,but the shortage of oil resources and the environmental pollution caused by the traditional internal combustion engine vehicles are also highlighted.Therefore,countries are committed to the development and promotion of new energy vehicles,new energy vehicles will become the future direction of the development of the automobile.But because the new energy vehicles are not powered by the engine to provide power,without the sound of the engine,so at low speed when the car is particularly quiet,so pedestrian is difficult to find and timely avoidance vehicles,easy to cause traffic accidents,while driving at high speed,due to the existence of tire noise and wind noise,there is no such problem.In this paper,combined with the specific problems of electric vehicles in low speed driving,we design and synthesize the warning sound samples by the method of speech waveform editing and synthesis,and get a lot of warning samples.According to the subjective evaluation of sound quality and the objective psychologic acoustics analysis method,the sound samples with the highest scores in the three speed ranges are selected as the warning sound,a evaluation and prediction model is established.On the basis of the selection of the warning sound,the hardware circuit design and software program development of the control system are completed,and through the road test to verify the warning sound warning effect and control system of the real-time and reliability.Firstly,acquisition of traditional internal combustion engine car outer sound,using speech synthesis technology,the amplitude of the sound signal is changed in a certain frequency range,a large number of sound signal samples can be obtained by changing the amplitude of the waveform in the frequency range.Secondly,based on the concept of sound quality,get all of the sound samples subjective evalution experiment,and calculate the objective psychoacoustic parameters of each sound signal,select the score of the best sound samples as the corresponding speed range warning sound.Based on the BP neural network optimized by simulated annealing algorithm and genetic algorithm,and the subjective and objective evaluation data,a electric vehicle warning sound quality assessment and forecast model was established,and the model has high prediction accuracy.The model can be used to replace the subjective evaluation when the warning sound is designed,save manpower and material resources and simplify the process of warning sound design.Finally,according to the actual automobile road driving condition,the hardware circuit design and the software control program of the electric vehicle warning sound control system are completed,and build up the complete control system.The road test of electric vehicle equipped with warning sound system is verified the designed warning sound to achieve the desired effect of pedestrian warning,at the same time,the reliability of the designed warning system is verified.