Search

Passenger thermal comfort in high-speed train (HST) carriages presents unique challenges due to factors such as extensive operational areas, longer travel durations, larger spaces, and higher passenger capacities. This study aims to propose a new prediction model to better understand and address thermal comfort in HST carriages. The proposed prediction model incorporates skin wettedness, vertical skin temperature difference (ΔTd), and skin temperature as parameters to predict the thermal sensation vote (TSV) of HST passengers. The experiments were conducted with 65 subjects, evenly distributed throughout the HST compartment. Thermal environmental conditions and physiological signals were measured to capture the subjects' thermal responses. The study also investigated regional and overall thermal sensations experienced by the subjects. Results revealed significant regional differences in skin temperature between upper and lower body parts. By analyzing data from 45 subjects, We analyzed the effect of 25 variables on TSV by partial least squares (PLS), from which we singled out 3 key factors. And the optimal multiple regression equation was derived to predict the TSV of HST occupants. Validation with an additional 20 subjects demonstrated a strong linear correlation (0.965) between the actual TSV and the predicted values, confirming the feasibility and accuracy of the developed prediction model. By integrating skin wettedness and ΔTd with skin temperature, the model provides a comprehensive approach to predicting thermal comfort in HST environments. This research contributes to advancing thermal comfort analysis in HST and offers valuable insights for optimizing HST system design and operation to meet passengers' comfort requirements. [ABSTRACT FROM AUTHOR]

The overall comfort of train passenger is influenced by many environmental factors such as vibration, noise and pressure. However, the couple effect of these influencing factors causes the difficulty in evaluating the overall comfort. This study revealed the potential comfort degradation mechanisms in high-speed railway environments and proposed a machine learning evaluation model to assess passenger comfort. Here, the subjective overall comfort ratings and the electroencephalography (EEG) of twenty passengers were collected in the field tests. Compared with passengers who were in a state of comfort, the brain areas (BA6/13/20/24/31/40/47) of passengers who felt uncomfortable were all significantly activated in the beta band. Based on the neural signature above, three related human reactions when passengers feel uncomfortable were recognized, including perceiving the environment, inducing negative emotions, and finally producing body movement intention. To assess the overall comfort of train passengers, six kinds of features extracted from EEG signals were used to train an evaluation model based on the LightGBM algorithm. This work offers a neurological explanation for the mechanisms of degradation of overall comfort and provides a novel and effective method to assess it. • Carried out a field test in high-speed railway operating environments. • Collect the EEG signals and questionnaires to assess passenger overall comfort. • Establish a machine learning model for overall comfort evaluation. • First-of-its-kind that reveal the discomfort mechanism based on neural signatures. • Prove that the emotional regulation has a great influence on passenger comfort. [ABSTRACT FROM AUTHOR]

Heating, Ventilation, and Air Conditioning (HVAC) systems in high-speed trains (HST) are responsible for consuming approximately 70% of non-operational energy sources, yet they frequently fail to ensure provide adequate thermal comfort for the majority of passengers. Recent advancements in portable wearable sensors have opened up new possibilities for real-time detection of occupant thermal comfort status and timely feedback to the HVAC system. However, since occupant thermal comfort is subjective and cannot be directly measured, it is generally inferred from thermal environment parameters or physiological signals of occupants within the HST compartment. This paper presents a field test conducted to assess the thermal comfort of occupants within HST compartments. Leveraging physiological signals, including skin temperature, galvanic skin reaction, heart rate, and ambient temperature, we propose a Predicted Thermal Comfort (PTC) model for HST cabin occupants and establish an intelligent regulation model for the HVAC system. Nine input factors, comprising physiological signals, individual physiological characteristics, compartment seating, and ambient temperature, were formulated for the PTS model. In order to obtain an efficient and accurate PTC prediction model for HST cabin occupants, we compared the accuracy of different subsets of features trained by Machine Learning (ML) models of Random Forest, Decision Tree, Vector Machine and K-neighbourhood. We divided all the predicted feature values into four subsets, and did hyperparameter optimisation for each ML model. The HST compartment occupant PTC prediction model trained by Random Forest model obtained 90.4% Accuracy (F1 macro = 0.889). Subsequent sensitivity analyses of the best predictive models were then performed using SHapley Additive explanation (SHAP) and data-based sensitivity analysis (DSA) methods. The development of a more accurate and operationally efficient thermal comfort prediction model for HST occupants allows for precise and detailed feedback to the HVAC system. Consequently, the HVAC system can make the most appropriate and effective air supply adjustments, leading to improved satisfaction rates for HST occupant thermal comfort and the avoidance of energy wastage caused by inaccurate and untimely predictive feedback. • This paper presents thermal comfort model for high-speed train passengers based on skin temperature, galvanic skin response and electrocardiogram. • Differences in physiological signals and thermal sensation votes among different populations (Gender, BMI and position) were analyzed. • The features of the four combinations were used as input factors to obtain accuracy using different machine learning models. • Sensitivity analysis of the best prediction model (Random Forest) was performed using SHAP and DSA methods. [ABSTRACT FROM AUTHOR]