Research On Evaluation And Regulation Of Rail Passenger Noise Annoyance Based On Physiological Indices

With the rapid development of social economy and the improvement of people’s life qualities,passengers are not only satisfied with the fast and safe way of travel,but also put forward higher requirements for the ride comfort of the train.Generally,the low-frequency characteristics of ordinary railway noise are not obvious.However,under the condition of high-speed operation,the long wave irregularity leads to the vibration of the car-body structure,which will further increase the low-frequency structural noise with lower natural frequency in the car-body.In the meanwhile,with the continuous improvement of the speed,the aerodynamic noise with significant low-frequency characteristics has become the main source of railway noise.The impact of low-frequency noise on human body is mainly manifested in the changes of EEG,ECG and other physiological factors.Long time exposure will cause passengers’ annoyance and reduce their ride comfort.Compared with medium and high frequency noise,low-frequency noise is not easy to be effectively controlled,which makes the low-frequency noise increasingly more significant.This will make passengers annoying and irritable under the influence of low-frequency noise when riding,thus reducing the ride comfort of passengers.A-weighting evaluation is one of the most widely used noise evaluating methods.However,A-weighting network cannot be applied to the evaluation of low-frequency noise due to its large attenuation and underestimation.The existing evaluation and analysis method of low-frequency noise in the carriage is not perfect.They only consider a single or a few factors,but ignore the individual physiology and psychology of passengers.Therefore,the evaluation method of low-frequency noise inside the train and the annoyance of train passengers exposed to low frequency noise need to be further studied.Therefore,it is of great engineering significance and application value to carry out the evaluation of low-frequency noise inside the train and the research on the passengers’ physiological noise annoyance index.In view of the key problems in the research of passenger annoyance under low-frequency noise exposure in the vehicle,this dissertation mainly carries out the following research work.1.The passenger’s annoyance evaluation method based on Tokita aviation low frequency noise index in train is proposed.Tokita aviation index is introduced to evaluate the sound pressure level and sound quality of the vehicle.The loudness,sharpness,roughness and psychoacoustics annoyance evaluation in the of high-speed trains and the metro railway at different speeds is used for verification.The results show that the proposed evaluation method of passenger annoyance of low-frequency noise can highlight the impact of low-frequency noise in train.2.The ECG and EEG denoising method based on singular value decompition and adaptive periodic segment matrix is proposed.The basic principle is: according to periodic characteristics of ECG and random characteristics of muscle artifacts,the rank 1 trajectory matrix containing ECG characteristic information is constructed,and the singular value decomposition is performed to obtain the maximum singular value.This value is used to reconstruct the ECG signal,which can eliminate the influence of muscle artifacts and ECG artifacts in the process of passengers’ ECG and EEG acquisition.Finally,the pure EEG and ECG signals are recovered.In the process of ECG denoising,the RR interval constraint is used to select the segment length,and the R peak and its position are used to construct a strict rank1 trajectory matrix.The first largest singular value is used to restore the pure ECG signal.Similarly,through the above method,the periodic ECG artifacts are extracted in the process of EEG,so as to achieve the effect of removing the ECG artifacts and extracting pure EEG.The effectiveness of the proposed method is verified by analyzing the data from different open EEG and ECG databases.3.The passengers’ noise annoyance indices based on passenger heart rate variability and EEG rhythm are proposed.According to the principle that heart rate and RR interval change when passengers exposed to noise,the annoyance index of ECG based on heart rate variability is proposed.Through the analysis of heart rate variability under noise exposure,the validity of the proposed index was verified.In the meanwhile,based on the feedback information of EEG in noise environment,the related representation and theoretical basis of EEG,a noise annoyance index of EEG based on multi-rhythm is proposed.The validity of the proposed index is verified by the EEG of the passengers exposed to interior noise in both metro railways and the high-speed trains at different speeds.4.Based on the theory of sound adjustment,the method of adding sound samples is proposed to improve interior noise and passenger comfort.The principle of this method is as follows: the low-frequency noise component in the original interior sound sample is significant.According to the concept of frequency balance in sound landscape,the original interior sound sample is superimposed with artifact and natural sounds,such as pink noise,FM pure tone,waterloo,bird chirp.The sound quality of the interior noise before and after superimposition and the annoyance of passengers are analyzed,so as to achieve an appropriate balance of high,medium and low frequencies to improve the interior sound quality and ride comfort.