Acoustic comfort assessment and regulation system suitable for mute carriage

Abstract

The invention relates to the technical field of acoustic regulation and control, in particular to an acoustic comfort level evaluation and regulation and control system suitable for a mute carriage, and the system comprises an evaluation system which comprises a data collection module, a data processing module, a data analysis module and an acoustic comfort level visualization module; the regulation and control system comprises a train compartment acoustic comfort visualization module, a global and local active noise reduction module and a mute compartment acoustic comfort alarm and regulation and control module; the system has the advantages that separation of various kinds of noise in the compartment is achieved through the audio separation technology, the train noise quality is evaluated in real time, non-train noise such as passenger flow noise and loudspeaker music in the compartment is monitored, physiological comfort evaluation of passengers is also considered, noise monitoring of the mute compartment is achieved in the true sense, and guidance is provided for a regulation and control system.

Description

Technical Field

[0001] The present invention relates to the field of acoustic control, and in particular to an acoustic comfort evaluation and control system suitable for a silent carriage. Background Art

[0002] With the rapid development of the high-speed railway industry, passengers are paying more and more attention to the noise problem in the carriage. With the continuous increase in the speed of trains, the aerodynamic noise with significant low-frequency characteristics and the low-frequency structural noise caused by long-wave irregularities in the noise inside the high-speed train are important factors affecting the riding comfort of passengers. During the operation of the train, there are many passengers in the carriage. In addition to railway noise such as wheel-rail noise, traction noise, and aerodynamic noise, some passengers make loud calls and children make noises, which will also affect the riding comfort of passengers. In particular, in recent years, in order to further improve the quality of high-speed rail operation services and optimize the travel experience of passengers, the railway department has provided silent carriage services on the Beijing-Shanghai, Chengdu-Chongqing, Xicheng and other lines since December 2020, and has opened the function of railway 12306 to choose to purchase silent carriage tickets on its own. However, the actual situation shows that the silent carriage is not silent. On the one hand, the noise energy has increased sharply due to the increase in operating speed and the influence of tunnel working conditions. On the other hand, some passengers lack self-awareness or moral sense, and it is difficult for crew members to take effective measures in time. Therefore, real-time evaluation of cabin acoustic comfort and proposing practical solutions are important guarantees for promoting the implementation of quiet cabins.

[0003] However, some current studies have proposed a series of sound quality evaluation models by collecting in-car noise samples and then evaluating them. Such models cannot directly reflect the acoustic comfort experience of passengers and ignore the individual differences in passengers' sensitivity to noise. Although some measures in silent cars are considerate from a service perspective, in practice, the effect is not satisfactory.

[0004] Therefore, it is urgent to establish an acoustic comfort evaluation and control system for passengers in quiet carriages, and use non-contact physiological and visual signal acquisition technology to identify the individual physiological comfort of passengers, so as to meet the personalized service needs of quiet carriages and improve the work efficiency of crew members. Summary of the invention

[0005] The purpose of the present invention is to provide an acoustic comfort evaluation and control system suitable for silent cars, aiming to improve the defect that the current silent cars are not silent, identify the noise type of the silent car through audio separation technology, evaluate the sound quality of various noises, combine the physiological and visual signals of the passengers to identify the individual physiological comfort, and realize the real-time monitoring of the acoustic comfort (sound quality, non-train noise sound pressure level, physiological comfort) of the silent car, and put forward corresponding control measures in a targeted manner, with active noise reduction devices and acoustic visualization functions, to realize the acoustic comfort control of the silent car and improve the work efficiency of the crew. The system proposed can meet the personalized service needs of the passengers and improve the service quality of the silent car. To solve the problems raised in the above background technology.

[0006] To achieve the above-mentioned object, the present invention provides the following technical solutions: an acoustic comfort evaluation and control system suitable for a silent car compartment, the system comprising: an evaluation system including a data acquisition module, a data processing module, a data analysis module, and an acoustic comfort visualization module; The control system includes a train compartment acoustic comfort visualization module, a global and local active noise reduction module, and a silent compartment acoustic comfort alarm and control module.

[0007] Preferably, the data acquisition module includes: a) Noise data collection: A noise sensor is used to obtain the noise signal inside the train. The noise collection period is 5 minutes after the train starts and 15 minutes before the train arrives at the station, and the noise is transmitted to the data processing module; b) Physiological signal acquisition: linear frequency modulation continuous wave is used to monitor the heart rate and breathing rate of the occupants; c) Visual signal acquisition: High-definition cameras are used to collect passengers’ facial expressions and body movements in real time.

[0008] Preferably, the data processing module includes: a) For noise data, regardless of the broadcasting time period, audio separation technology is used to separate different types of sounds such as train noise, pedestrian noise, and external music. Appropriate time window length is set for noise to divide the segments, and a suitable weighted window function is selected to ensure that the noise signal meets the periodic requirements of fast Fourier transform (FFT) processing, eliminate high-frequency interference and leakage energy, and use auditory filters to divide the frequency bands. Since train noise has significant low-frequency characteristics, the A-weighted sound pressure level is not suitable for evaluating train noise. Therefore, the loudness, sharpness, roughness, and jitter psychoacoustic parameters of the current noise signal are calculated based on spectrum analysis, and the trained BP neural network model is used to obtain the sound quality discomfort score and normalize it; for pedestrian noise and external music non-train noise, the A-weighted sound pressure level of the current carriage non-train noise is calculated in real time; b) For physiological signals, the radar echo signal is adaptively decomposed into the form of adding multiple intrinsic mode function (IMF) components, effectively distinguishing and extracting breathing and heartbeat signals, and obtaining the heart rate variability, low-frequency and high-frequency power, and breathing rate biomarkers of the occupants through feature extraction. In order to reduce the signal feature differences caused by individual variability, the average change rate of each feature is calculated by comparing with the baseline value; c) For visual signals, deep learning methods are used to extract the facial expression features and spatiotemporal features of the occupants’ body movements to assist in determining the occupants’ acoustic comfort experience.

[0009] Preferably, the data analysis module includes: The sound quality discomfort score is divided into two states: discomfort and comfort; the A-weighted sound pressure level thresholds of pedestrian noise and external music non-train noise are set; so as to timely understand the noise conditions in the carriage; the extracted physiological and visual signal features are input into the machine learning model, and linear discriminant analysis is used to project the original samples into a low-dimensional sample space. At the same time, the best feature set with a strong correlation with noise discomfort is selected, and a classification model is established to evaluate the physiological comfort of the occupants.

[0010] Preferably, the acoustic comfort visualization module includes: a) Presentation of train noise quality status; b) Numerical presentation of A-weighted sound pressure level of non-train noise; c) Occupant physiological comfort: Taking into account individual noise sensitivity, the occupant's physiological signals are used as the degree of noise discomfort in a small range, that is, the individual noise discomfort assessment standard. According to the evaluation results of the classification model, noise discomfort is divided into relatively quiet, slightly noisy, and very noisy, and the occupant's noise discomfort status is presented through a visualization platform.

[0011] Preferably, the train compartment acoustic comfort visualization module includes: A data acquisition module is used to input the sound quality status, non-train noise sound pressure level, and occupant physiological comfort into the visualization platform. Based on acoustic holography and monocular vision technology, the acoustic visualization inside the carriage is realized and uploaded to the visualization platform simultaneously, and presented to the crew intuitively, helping them to quickly grasp the main causes of carriage noise discomfort and the approximate location of the sound source.

[0012] Preferably, the global and local active noise reduction modules include: The system uses real-time calculation of the cabin noise discomfort score and individual comfort status assessment. If the current noise discomfort score is too high or the noise discomfort status of more than 80% of the passengers in the car is "very noisy", the active noise reduction function will be turned on through the three speakers at the front, middle and rear of the cabin. Conversely, if the noise discomfort status of some passengers is "very noisy", the visualization platform will remind the passengers to turn on the local active noise reduction function through the speakers in the seat area.

[0013] Preferably, the silent compartment acoustic comfort alarm and control module includes three judgment conditions: First, when the system detects that the sound pressure level of passenger noise and external music noise in the car exceeds the calibration threshold; second, the current sound quality state of the quiet car is uncomfortable or the physiological comfort level of 80% of the passengers is very noisy; third, none of the above conditions are met, but the physiological comfort level of some passengers is very noisy.

[0014] Preferably, when only condition one is met, the visualization page will remind the crew member to go to the quiet compartment with a yellow flashing alarm and intervene according to the acoustic visualization results; when only condition two is met, the visualization page will remind the crew member to turn on the global active noise reduction function in the compartment with a blue flashing alarm; if both conditions one and two are met, the visualization page will remind the crew member to turn on the global active noise reduction function and go to the compartment with a red flashing alarm; when only condition three is met, the local active noise reduction function of the seat where the occupant is located is activated, and the crew member is also reminded with a yellow flashing alarm, combined with the analysis of the occupant's facial expression and body movements, to go to the quiet compartment and intervene as appropriate; each alarm is accompanied by a voice reminder, and is used in conjunction with the high-definition camera in the compartment to identify whether the occupant has abnormal behavior or movement, to ensure that the crew member is informed in a timely manner and takes appropriate measures to ensure the acoustic comfort of the quiet compartment.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The acoustic comfort evaluation and control system for silent cars proposed in the present invention achieves the separation of various types of noise in the car through audio separation technology. It not only evaluates the sound quality of train noise in real time, and monitors non-train noise such as pedestrian noise and external music in the car, but also takes into account the physiological comfort evaluation of passengers, thus truly realizing noise monitoring of silent cars and providing guidance for the control system. In the control system, global and local active noise reduction measures are proposed according to the sound quality status, non-train noise sound pressure level, and physiological comfort of passengers, and acoustic visualization is used to help crew members quickly identify the approximate location of non-train noise such as pedestrian noise and external music, so as to intervene in time. Compared with previous acoustic comfort evaluation methods, the present invention takes into account the impact of non-train noise, provides technical support for personalized services in silent cars, and is expected to improve the overall satisfaction of passengers with silent cars. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a system block diagram of the present invention; Figure 2 It is a flow chart of the system of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose and technical solution of the present invention clearly and completely described, and the advantages more clearly understood, the embodiments of the present invention are further described in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are part of the embodiments of the present invention, rather than all of the embodiments, and are only used to explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention. All other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1, refer to the attached Figure 1 The present invention provides a technical solution: an acoustic comfort evaluation system suitable for a silent car compartment: 1) Data collection a) Noise signal acquisition Noise sensors are installed at appropriate positions in the front, middle and rear of the train compartment. 5 minutes after the train starts, noise signals at the front, middle and rear of the compartment are collected. The noise collection period is 5 minutes after the train starts and 15 minutes before the train arrives at the station, and the collected noise signals are transmitted to the data processing module.

[0018] b) Physiological signal collection The R60ABD1 millimeter-wave radar sensor can be installed in an appropriate position in the car. It adopts 60 GHz millimeter-wave radar technology and is based on the linear frequency-modulated continuous wave (FMCW) radar system. It can accurately scan the motion tomography of the human body and monitor the breathing and heart rate of the occupants in different postures.

[0019] c) Visual signal acquisition Install high-definition cameras in train carriages to obtain facial image information of passengers.

[0020] 2) Data processing module ①Noise data processing a) Audio signal separation: Use the trained AudioSep – separate anything audio model to separate various noise signals such as train noise, pedestrian noise, and mobile phone music.

[0021] b) Obtaining evaluation indicators: Set appropriate time window lengths for the noise signal to divide it into segments, perform Hanning window processing on the non-periodic continuous noise signal to ensure that the noise signal meets the periodic requirements of fast Fourier transform (FFT) processing, eliminate high-frequency interference and leakage energy, and use a one-third octave band filter for spectrum analysis. Since the low-frequency energy of train noise is prominent, the A-weighted sound pressure level in the existing standard cannot accurately evaluate train noise. Therefore, psychoacoustic parameters such as loudness, sharpness, roughness, and jitter are used and calculated in real time. The trained BP neural network model is used to obtain the car sound quality discomfort score and normalize it; for non-train noise such as pedestrian noise and external music, the autocorrelation spectrum of various noise sound pressures is obtained, and the sound pressure level (dBA) of various noises is calculated.

[0022] ②Physiological signal processing The local cutoff frequency of the adaptive signal using time varying filtering based on EMD (TVF-EMD) can effectively improve the signal separation performance. The intrinsic mode functions (IMF) components decomposed by TVF-EMD are used to reconstruct the time domain signal corresponding to the heartbeat, and the occupant's heart rate mean (HR), the mean of the normal RR interval, the standard deviation of the average normal RR interval every 5 minutes (SDNN), low-frequency power, high-frequency power, skin conductivity (SCL) and other biomarkers are obtained through feature extraction. In order to reduce the difference in signal features caused by individual variability, the average change rate of each feature is calculated by comparing with the baseline value.

[0023]

[0024] Note: is the indicator baseline value, is the index value at the i-th moment.

[0025] ③Visual signal processing The VGG-19 network is used to extract the facial expression features of the occupants, and the ResNet50 network is used to extract the facial angle features. In terms of motion feature extraction, a method of human body region fusion is proposed, and Yolov3 and C3D networks are used to extract the spatiotemporal features of limb movements.

[0026] 3) Data analysis module ① Based on the score of discomfort of sound quality, it is divided into two states: discomfort and comfort.

[0027] ② At the same time, a threshold is set to measure the A-weighted sound pressure level of non-train noise such as pedestrian noise and external music, so that the noise level in the car can be monitored in real time.

[0028] ③ Using machine learning models, these physiological and visual signals can be processed to accurately distinguish the physiological comfort of occupants, and effectively classify the physiological comfort of occupants into three categories (relatively quiet, somewhat noisy, and very noisy), providing a stable and reliable way to predict and evaluate the physiological comfort of occupants.

[0029] The specific implementation process of the three-class Softmax regression model is as follows: The selected strongly correlated physiological and visual features, a total of n, are used as the input vector , linear transformation is used to obtain three physiological comfort categories ( =1 relatively quiet, =2Somewhat noisy, = 3 very noisy), and the category score Convert to probability Finally, the category with the largest probability is selected as the prediction result of the occupant's physiological comfort. The cross entropy loss function is used to measure the gap between the probability distribution of the model output and the actual label, and the gradient descent method is used to minimize the loss function and update the weights. and bias

[0030]

[0031] Note: Yes Category The score, is the weight parameter, is the bias term.

[0032]

[0033] Note: Represents input belong Probability of physiological comfort.

[0034] 4) Acoustic comfort visualization module a) Presentation of train noise quality status; b) Numerical presentation of A-weighted sound pressure level of non-train noise; c) Occupant physiological comfort: Taking into account individual noise sensitivity, the occupant's physiological signals are used as the degree of noise discomfort in a small range, that is, the individual noise discomfort assessment standard. According to the evaluation results of the classification model, noise discomfort is divided into relatively quiet, slightly noisy, and very noisy, and the occupant's noise discomfort status is presented through a visualization platform.

[0035] 5) Acoustic comfort assessment module The acoustic comfort of the silent carriage is evaluated from three aspects: sound quality status, non-train noise sound pressure level, and passenger physiological comfort. Specifically, the sound quality status is based on the sound quality being in an uncomfortable state, the non-train noise sound pressure level being set to exceed the calibration threshold (such as 68dBA), and the passenger physiological comfort being "very noisy". If any of these conditions are met, the acoustic comfort requirements are not met and further acoustic comfort control is required to provide input for the acoustic comfort control system.

[0036] Figure 1 An acoustic comfort evaluation system suitable for silent carriages is demonstrated, through which the acoustic comfort of silent carriages can be monitored in real time. First, the noise signal in the car and the physiological signal of the passengers are obtained through noise sensors and multimodal physiological acquisition bracelets; secondly, various types of noise are identified through audio separation technology, and the sound quality discomfort score of train noise and the A-weighted sound pressure level of non-train noise are calculated. At the same time, the breathing and heartbeat signals of the passengers are extracted for feature extraction; then, the thresholds of the sound quality discomfort score and the A-weighted sound pressure level are set, and a physiological comfort evaluation model for the members is established; finally, the acoustic comfort of the carriage is visualized.

[0037] Embodiment 2, refer to the attached Figure 2 , based on the first embodiment, an acoustic comfort control system suitable for a silent car is proposed.

[0038] (1) Visualization of acoustic comfort in quiet cabins The sound quality status, non-train noise sound pressure level, and passenger physiological comfort information obtained from the acoustic comfort assessment are presented to the crew through a visualization platform. When the sound quality status is uncomfortable or the non-train noise sound pressure level is greater than the threshold, or the passenger physiological comfort is very noisy, it is presented in red eye-catching fonts. Through the in-car noise sensor and high-definition camera, the acoustic visualization of non-train noise in the car is achieved based on acoustic holography and monocular vision technology, which makes it easy for crew members to quickly find the approximate noise area and intervene.

[0039] (2) Active noise reduction device for quiet cabin a) Global active noise reduction device The global active noise reduction device includes a feedforward signal collector, a control processor, a noise sensor, and a speaker. The ambient noise signal and the feedforward signal are obtained through the noise sensor and the feedforward signal collector; the control processor corrects the generated active noise reduction signal in combination with the feedforward signal; and finally the corrected active noise reduction signal is output through the speaker. The speakers are arranged at both ends of the car to give full play to the radiation coverage effect of the sealed large area active noise reduction technology. When the current sound quality state of the silent car is uncomfortable or the physiological comfort of 80% of the passengers is very noisy, turn on the global active noise reduction device.

[0040] b) Local active noise reduction device The local active noise reduction device only adds more speakers based on a), and is arranged at the ear height of the passenger seat. When the silent cabin does not currently meet the conditions for turning on the global active noise reduction device, but the physiological comfort of some passengers is still very noisy, the system activates the local active noise reduction device in the area where the passenger is located.

[0041] (3) Early warning and control of acoustic comfort in quiet cabins The judgment conditions and control measures are as follows: Condition 1: When the system detects that the sound pressure level of noise such as passenger noise and external music in the carriage exceeds the calibration threshold, the noise limit of 68dB(A) in the middle of the second-class carriage in TJ / CL 342A-2014 "Provisional Technical Conditions for Chinese Standard EMUs with a Speed ​​of 350 km / h" is used as the threshold. The specific implementation threshold can be adjusted according to the noise limit specified in the relevant standards of different trains.

[0042] Specific implementable control measures: The visualization page will remind the crew to go to the quiet car with a yellow flashing alarm and intervene based on the acoustic visualization results. Generally, this situation is manifested by some passengers not complying with the relevant regulations of the quiet car and creating additional noise. In order to avoid conflicts between passengers, the crew needs to mediate or take tough measures according to the prompts.

[0043] Condition 2: The current sound quality state of the quiet cabin is uncomfortable or the physiological comfort level of 80% of the passengers is very noisy.

[0044] Specific implementable control measures: The visualization page uses a blue flashing alarm to remind crew members to turn on the global active noise reduction function in the carriage to ensure the overall acoustic comfort of the train's quiet carriages.

[0045] Condition 3: Both conditions 1 and 2 are not met, that is, the sound pressure level of non-train noise does not exceed the threshold, and the sound quality state is comfortable, but the physiological comfort of some passengers is very noisy.

[0046] Specific control measures that can be implemented: Activate the local active noise reduction function of the seat where the passenger is located, and use a yellow flashing alarm to remind the crew to go to the quiet compartment to intervene as appropriate. Since conditions 1 and 2 are not met and there are no other passengers making additional noise, this type of passenger is noise-sensitive and earplugs can be provided according to their needs.

[0047] Note: If conditions 1 and 2 are both met, the visualization page will use a red flashing alarm to remind the crew to turn on the global active noise reduction function and go to the car to intervene and adjust the function.

[0048] Figure 2 A schematic diagram of acoustic comfort control for a quiet cabin is shown. Figure 1The acoustic comfort evaluation system shown first inputs the current cabin sound quality discomfort state, counts the proportion of physiological comfort of all passengers, and inputs the non-train noise sound pressure level (default A-weighted); when the sound quality discomfort state is uncomfortable or the physiological comfort of more than 80% of the passengers is "very noisy", measure 1 is implemented: the global active noise reduction function of the cabin is automatically turned on, otherwise it does not need to be turned on. If the physiological comfort of some passengers is "very noisy", measure 2 is implemented: the local active noise reduction function of the passenger seat is activated; when the A-weighted sound pressure level of train noise is greater than 68dBA, measure 3 is implemented: the crew goes to the cabin to intervene according to the prompt of the alarm system; after the above operations are completed, continuous monitoring or continued regulation is carried out according to the actual state evaluation of the acoustic comfort of the silent cabin.

[0049] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An acoustic comfort evaluation and control system suitable for a silent car, characterized in that: The system comprises: an evaluation system including a data acquisition module, a data processing module, a data analysis module, and an acoustic comfort visualization module; The control system includes a train compartment acoustic comfort visualization module, a global and local active noise reduction module, and a silent compartment acoustic comfort alarm and control module.

2. The acoustic comfort evaluation and control system for a silent car compartment according to claim 1, characterized in that: The data acquisition module comprises: a) Noise data collection: A noise sensor is used to obtain the noise signal inside the train. The noise collection period is 5 minutes after the train starts and 15 minutes before the train arrives at the station, and the noise is transmitted to the data processing module; b) Physiological signal acquisition: linear frequency modulation continuous wave is used to monitor the heart rate and breathing rate of the occupants; c) Visual signal acquisition: High-definition cameras are used to collect passengers’ facial expressions and body movements in real time.

3. The acoustic comfort evaluation and control system for a silent car compartment according to claim 1, characterized in that: The data processing module comprises: a) For noise data, regardless of the broadcasting time period, audio separation technology is used to separate different types of sounds such as train noise, pedestrian noise, and external music. Appropriate time window length is set for noise to divide the segments, and a suitable weighted window function is selected to ensure that the noise signal meets the periodic requirements of fast Fourier transform (FFT) processing, eliminate high-frequency interference and leakage energy, and use auditory filters to divide the frequency bands. Since train noise has significant low-frequency characteristics, the A-weighted sound pressure level is not suitable for evaluating train noise. Therefore, the loudness, sharpness, roughness, and jitter psychoacoustic parameters of the current noise signal are calculated based on spectrum analysis, and the trained BP neural network model is used to obtain the sound quality discomfort score and normalize it; for pedestrian noise and external music non-train noise, the A-weighted sound pressure level of the current carriage non-train noise is calculated in real time; b) For physiological signals, the radar echo signal is adaptively decomposed into the form of adding multiple intrinsic mode function (IMF) components, effectively distinguishing and extracting breathing and heartbeat signals, and obtaining the heart rate variability, low-frequency and high-frequency power, and breathing rate biomarkers of the occupants through feature extraction. In order to reduce the signal feature differences caused by individual variability, the average change rate of each feature is calculated by comparing with the baseline value; c) For visual signals, deep learning methods are used to extract the facial expression features and spatiotemporal features of the occupants’ body movements to assist in determining the occupants’ acoustic comfort experience.

4. The acoustic comfort evaluation and control system for a silent car according to claim 1, characterized in that: The data analysis module includes: The sound quality discomfort score is divided into two states: discomfort and comfort; the A-weighted sound pressure level thresholds of pedestrian noise and external music non-train noise are set; so as to timely understand the noise conditions in the carriage; the extracted physiological and visual signal features are input into the machine learning model, and linear discriminant analysis is used to project the original samples into a low-dimensional sample space. At the same time, the best feature set with a strong correlation with noise discomfort is selected, and a classification model is established to evaluate the physiological comfort of the occupants.

5. The acoustic comfort evaluation and control system for a silent car compartment according to claim 1, characterized in that: The acoustic comfort visualization module includes: a) Presentation of train noise quality status; b) Numerical presentation of A-weighted sound pressure level of non-train noise; c) Occupant physiological comfort: Taking into account individual noise sensitivity, the occupant's physiological signals are used as the degree of noise discomfort in a small range, that is, the individual noise discomfort assessment standard. According to the evaluation results of the classification model, noise discomfort is divided into relatively quiet, slightly noisy, and very noisy, and the occupant's noise discomfort status is presented through a visualization platform.

6. The acoustic comfort evaluation and control system for a silent car compartment according to claim 1, characterized in that: The train compartment acoustic comfort visualization module comprises: A data acquisition module is used to input the sound quality status, non-train noise sound pressure level, and occupant physiological comfort into the visualization platform. Based on acoustic holography and monocular vision technology, the acoustic visualization inside the carriage is realized and uploaded to the visualization platform simultaneously, and presented to the crew intuitively, helping them to quickly grasp the main causes of carriage noise discomfort and the approximate location of the sound source.

7. The acoustic comfort evaluation and control system for a silent car according to claim 1, characterized in that: The global and local active noise reduction modules include: The cabin noise discomfort score and individual comfort status are evaluated based on real-time calculation. If the current noise discomfort score is too high or the noise discomfort status of more than 80% of the passengers in the car is "very noisy", the active noise reduction function will be turned on through the three speakers at the front, middle and rear of the cabin. Conversely, if the noise discomfort status of some passengers is "very noisy", the visualization platform will remind the passengers to turn on the local active noise reduction function through the speakers in the seat area.

8. The acoustic comfort evaluation and control system for a silent car compartment according to claim 1, characterized in that: The silent compartment acoustic comfort alarm and control module includes three judgment conditions: First, when the system detects that the sound pressure level of passenger noise and external music noise in the car exceeds the calibration threshold; second, the current sound quality state of the quiet car is uncomfortable or the physiological comfort level of 80% of the passengers is very noisy; third, none of the above conditions are met, but the physiological comfort level of some passengers is very noisy.

9. The acoustic comfort evaluation and control system for a silent car compartment according to claim 8, characterized in that: When only condition one is met, the visualization page will remind the crew to go to the quiet compartment with a yellow flashing alarm and intervene according to the acoustic visualization results; when only condition two is met, the visualization page will remind the crew to turn on the global active noise reduction function in the compartment with a blue flashing alarm; if both conditions one and two are met, the visualization page will remind the crew to turn on the global active noise reduction function and go to the compartment with a red flashing alarm; when only condition three is met, the local active noise reduction function of the seat where the occupant is located is activated, and the crew is also reminded with a yellow flashing alarm, combined with the analysis of the occupant's facial expression and body movements, to go to the quiet compartment and intervene according to the situation; each alarm is accompanied by a voice reminder, and the high-definition camera in the compartment identifies whether the occupant has abnormal behavior or movement, ensuring that the crew is informed in time and takes appropriate measures to ensure the acoustic comfort of the quiet compartment.

Images