Hsuehshan Tunnel Research Articles

Data-driven numerical simulation with extended Kalman filtering and long short-term memory networks for highway traffic flow prediction

Abstract Developing an accurate and reliable computational tool for traffic flow prediction has always been an active research topic in transportation engineering and planning. The available predictive tools generally fall into parametric, nonparametric, and PDE-based approaches. In particular, the machine learning methods, such as the long short-term memory (LSTM) networks, belong to the nonparametric methods. This study proposes the data assimilation technique with LSTM for predicting highway traffic flows. The proposed method is developed under the framework of the extended Kalman filter (EKF) algorithm, which consists of two key components: the analysis and prediction steps. As the numerical simulator, a kernel component of the predictive tool, we use an explicit (EX) Godunov's scheme to discretize the Lighthill-Whitham- Richards model, where the MacNicholas formulation is used as the fundamental relation between the velocity and density. EKF combines LSTM prediction from two perspectives. In practical scenarios, future data at the upstream or downstream boundary points are unavailable. Therefore, the predicted values generated by LSTM are employed to set boundary conditions. Furthermore, two stages in EKF assimilate the LSTM predicted values, known as pseudo-observations, and the observed data in order with background values obtained through numerical simulation and observed data whenever available. This assimilation process aims to obtain a better initial condition for subsequent predictions, resulting in improved accuracy. Based on traffic data for the historical Hsuehshan Tunnel highway traffic data in Taiwan, the numerical results demonstrate that our method can effectively reduce the observation error and outperforms three baselines: EX, EKF, and LSTM.

Simulation and measurement of air quality in the traffic congestion area

The traffic congestion in the Hsuehshan tunnel and at the Toucheng interchange has led to traffic-related air pollution with increasing concern. To ensure the authenticity of our simulation, the concentration of the last 150 m in Hsuehshan tunnel was simulated using the computational fluid dynamics fluid model. The air quality at the Toucheng interchange along a 2 km length highway was simulated using the California Line Source Dispersion Model. The differences in air quality between rush hours and normal traffic conditions were also investigated. An unmanned aerial vehicle (UAV) with installed PM2.5 sensors was developed to obtain the three-dimensional distribution of pollutants. On different roads, during the weekend, the concentrations of pollutants such as SOx, CO, NO, and PM2.5 were observed to be in the range of 0.003–0.008, 7.5–15, 1.5–2.5 ppm, and 40–80 μg m− 3, respectively. On weekdays, the vehicle speed and the natural wind were 60 km h− 1 and 2.0 m s− 1, respectively. On weekdays, the SOx, CO, NO, and PM2.5 concentrations were found to be in the range of 0.002–0.003, 3–9, 0.7–1.8 ppm, and 35–50 μg m− 3, respectively. The UAV was used to verify that the PM2.5 concentrations of vertical changes at heights of 9.0, 7.0, 5.0, and 3.0 m were 45–48, 30–35, 25–30, and 50–52 μg m− 3, respectively. In addition, the predicted PM2.5 concentrations were 40–45, 25–30, 45–48, and 45–50 μg m− 3 on weekdays. These results provide a reference model for environmental impact assessments of long tunnels and traffic jam-prone areas. These models and data are useful for transportation planners in the context of creating traffic management plans.

Investigation of bus evacuation flow rates for tunnel fire quantitative risk assessment

ABSTRACT To increase the knowledge of bus evacuation in tunnel fire scenarios, we investigated the bus evacuation flow rate in the event of a tunnel fire, starting with a video analysis of the 2012 Hsuehshan tunnel fire in Taiwan, to gain a preliminary understanding of an actual bus evacuation flow rate. Two evacuation situations were manifested with different flow rates: the first one in which bus passengers non-urgently evacuated as in the case of a daily bus experience (the first seven evacuees, 0.28 PPS (Person Per Second)) and another situation in which passengers feel the urgency to evacuate (the next twelve evacuees, 0.67 PPS). Based on the investigation of the real fire incident, we conducted a bus evacuation experiment and a bus alighting observation. And the results of flow rates of normal exit were within 0.69–0.88 and 0.21–0.48 PPS, respectively. These results were similar to the actual flow rates in the two assumed evacuation situations in the Hsuehshan tunnel fire incident (nonurgent and urgent), respectively. Our approach can provide a viable approach to describe evacuation flow rates during tunnel fires. Moreover, using the flow rate of the normal alighting situation was preferable for highly reliable tunnel fire safety evaluation.

Role of Vehicular Catalytic Converter Temperature in Emission of Pollutants: An Assessment Based on Isotopic Analysis of CO2 and N2O.

Vehicular catalytic converters are used to regulate, reduce, and convert toxic and environmentally unfriendly compounds in exhaust gases into relatively inert and less harmful chemical species. The efficiency, however, is largely affected by the operating temperature of the converter which is set by the hot exhaust gas released from the combustion chamber. A major gas released during combustion is CO2, and its multiply substituted isotopocule, namely, 13C16O18O, provides a window of opportunity to probe directly the effective temperature of the converter in operation. Here, we report multiple isotopic measurements in exhaust CO2 (δ13C, δ17O, δ18O, and Δ47) of diesel (trucks and buses) and gasoline (sedans, trucks, and two-wheel motorcycles)-powered vehicles. For investigating the efficiency of a converter in reducing toxic compounds, we studied NOx processes through isotopic analysis of the exhaust N2O. We found that the degree of N2O reduction to N2 in gasoline-powered vehicles is high when the temperature is above 200 °C (inferred by Δ47). In contrast, diesel-powered vehicles produce N2O in abundance, probably a consequence of selective catalytic reduction of NOx, and the reduction efficiency depends on the converter temperature. In other words, the catalytic converters act as sinks and sources of N2O to the atmosphere in gasoline- and diesel-operated vehicles, respectively. We also report a new set of field data by measuring the isotopic compositions of CO2 and N2O in the Hsuehshan tunnel, a ∼13 km long highway tunnel in Taiwan. Elevated N2O concentrations inside the tunnel indicate that the emission of N2O by heavy-duty diesel vehicles is much higher compared to the reduction by gasoline-operated passenger cars, making the vehicular exhausts a net source of N2O to the atmosphere. The combined study of clumped isotopes and N2O concentration in exhaust gases suggests that it is useful to probe the operational temperature of catalytic converters and monitor the pollution level in operation, thus providing an opportunity for manufacturers to optimize the catalytic efficiency to reduce the level of toxic pollutants to the environment.

Switching Behavior and Control Policy of Congestion: Examples from Taiwan Highway System

The highway system is an important component of the modern society as it allows an efficient intercity communication. However, with the increase in the number of cars, the traffic congestion becomes a widespread problem around the world. Here, we study the traffic conditions in Hsuehshan tunnel in Taiwan. When there is traffic congestion, there is a remarkable fraction of vehicles choosing an alternative path, even if it cost more time than jamming on the highway. In addition, we evaluate through statistics and the government’s policy to improve the road condition of the tunnel.

Environmental impact of Hsuehshan Tunnel on water quality at Feitsui Reservoir and its tributaries.

To investigate the possible impact of the traffic flow of mountainous roads and the construction and operation of a long tunnel on the water quality of a reservoir, this study conducts statistical analysis on water quality, meteorological, and traffic data of the Feitsui Reservoir and its upstream tributaries over the last three decades. Results from statistical regression analysis indicate that in the upstream area where the traffic flow is low, water quality varies insignificantly with rainfall and traffic flow, providing as a background reference of natural environment. Water quality near a conventional highway on which many vehicles travel through the catchment area is significantly affected by both rainfall and traffic flow since the drainage system of a conventional highway sends its gathered water into adjacent rivers. Not only does traffic flow generate contaminants, but also the construction of the Hsuehshan Tunnel of the No. 5 Expressway, Taiwan, in the catchment area of the Feitsui Reservoir generates pollution. Drainage, silt settling and retarding basin, and wastewater treatment facilities near the construction site mitigate the impact of tunnel construction and traffic flow on the environment. The No. 5 Expressway makes good use of viaduct and tunnel structures, collecting water from pavements within the catchment area into sewage facilities, filtering it, and then emitting it outside the catchment area, forming a closed system over the Feitsui Reservoir. The Expressway now shortens travel time from two hours to 40min and accommodates 7-13 times previous traffic flows, insignificant influencing water quality in the upstream tributaries of the reservoir, demonstrating the effectiveness of its environmental protection measures.

Time gap distribution of bus alighting in tunnel fires

To understand time gaps during bus evacuation to conduct high-quality risk analysis of tunnel fires, this study collected time gaps from the Hsuehshan tunnel fire in Taiwan, experimental bus evacuations, and observed bus alighting. Based on the analysis of Hsuehshan tunnel fire, two situations were manifested: one wherein the bus passengers alighted normally as if it was a usual bus experience with no urgent requirement to evacuate (evacuees one to seven) and another associated with an urgency to evacuate (evacuees 8–19). We applied the results of the analysis in the Hsuehshan tunnel fire incident to investigate the time gaps from experiments and observations by dividing it into two groups: non-urgent and urgent.Statistical analyses were performed to investigate suitable probability distribution curves to describe the time gap distributions. The lognormal and gamma distributions show relatively good performance in approximating time gap data in non-urgent and urgent situations. Moreover, when comparing the time gaps of bus evacuation during the Hsuehshan tunnel fire accident, the urgent and non-urgent group time gaps corresponded to the assorted Stage II (52%) and Stage I (65%) of the Hsuehshan tunnel fire, respectively.

The hybrid lighting system with natural light and LED for tunnel lighting

The traditional tunnel lighting has the problem of poor lighting environment and huge energy consumption. This study uses natural light and LED as the light source to design two different lenses. When the natural light is sufficient, the natural light illumination system is used in the threshold zone and the exit zone, give tunnel lighting to reduce the number of LEDs used. When the natural light is insufficient, the sensor will detect the natural light level of the collimator exit to control the LED to compensate for the light or to illuminate the tunnel alone. In the two designs, the linear Fresnel lens is established by the Fresnel structure, and the free-form asymmetric lens of the LED is established by the catadioptric method. Finally, the model of the Hsueh-Shan tunnel was established by DIALux, and the two lenses were used simultaneously for mixed illumination to simulate average luminance, total uniformity, and longitudinal uniformity. The lens design reduces the influence of glare, and the design of the lamp spacing reduces the zebra pattern on the road surface. Combined with natural light and artificial light source, the tunnel light does not need to be 24 h of operation can save energy.

Analysis of smoke movement in the Hsuehshan tunnel fire

In 2012, a serious fire accident occurred in the Hsuehshan southbound road tunnel, which resulted in two fire fatalities and 34 injuries. A pickup truck was hit by a bus from behind; gasoline leaked from the bus and started the fire. Smoke was diverted from the southbound tunnel into the service tunnel and the northbound tunnel via two connection tunnels. Fire dynamics simulator (FDS) software was used to simulate the smoke movement in the 13 km long southbound road tunnel, employing more than 21 million computational cells. The heat release rate curve used in the simulation was based on a single bus fire which reached a maximum value of 30 MW. The smoke front movement obtained from the computer simulation was compared with that determined from the closed circuit TV (CCTV) recordings, and the agreement between the simulation results and CCTV recordings was sound. The critical velocity for controlling backlayering of a 30 MW bus fire was also determined from simulation, which agreed well with the value estimated from empirical correlations of model experiments. This research showed that during a fire, doors to connection tunnels should close automatically, and the service tunnel should only be used for emergency rescue purpose, instead of evacuation of people.

Theory Establishment and Data Preparedness for Modeling Emergency Medical Service in case of a Mass Casualty Incidents in Road Tunnels

Abstract The road system in Taiwan has expanded with remarkable speed resulting in great concern regarding the response to a mass casualty incident (MCI) by the Emergency Medical Services (EMSs). Accidents in road tunnels have the potential to cause major casualties due to their particular configurations and confined conditions. To meet this sudden demands involving huge uncertainties and complexities, EMSs provided by fire services need to be better prepared and more efficient. The recent research into EMS response to MCI aims at identifying effectiveness of life preservation and strategies of emergency mobility through outcome indicators to evaluate performance for different aspects of disaster management. However, the conducting of experimental studies is either impossible or ethically inappropriate. This study tries to use realistic injury statistics for EMS simulations based on medical records of a previous case study and an EMS database currently used in Taiwan. These statistics are temporarily used in a proposed trial MCI/EMS model where injury entities evolve in the medical response model, i.e. the model focuses only on the prehospital phase which includes triage, treatments at the scene and transportation processes - to understand the utilization of resources including Emergency Medical Technicians and ambulances. The outcome indicators also highlight the length of waiting including both before treatment at the Medical Post and before departure at the Delivery Post for each serious injury, and the total operation time to review the emergency response plan for an incident in Hsuehshan tunnel. This study also illustrates how such a model could be used to assess the impact on resource availability, implemented prioritization rule and various other strategies relating to emergency management.

Analysis of smoke control models at long tunnel: A case study on the Hsueh-shan tunnel

This study used the Fire Dynamics Simulator (FDS) to simulate a fire in a 5000 meters long section of the Hsuehshan tunnel (12.9km in total length). Actual distances were considered in jet fan simulations and steady-state velocity profiles were simulated in tunnel portals. A worst-case analysis was performed for the area between two vertical shafts in the tunnel. To prevent the smoke from becoming an obstacle in the Hsuehshan tunnel, fans located 250 meters upwind and 500 meters downwind of the fire source were deactivated. The smoke control system currently used in the Hsuehshan tunnel includes activation of four jet fans upwind of the fire and intermittent operation of jet fans downwind of the fire. This system was examined in this study as benchmark model Eva1 and compared with the alternative evacuation plans Eva2, Eva3 and Eva4. The results indicated that, regardless of fan activation, the downwind area of the fire remained dangerous. With regard to upwind safety, it was calculated that adults with an average walking speed of 1.2m/s and weak evacuees with an average walking speed of 0.64m/s would need 205 and 227 seconds respectively to traverse 30 meters. In Eva1 and Eva2, for 15MW, 30MW and 65MW heat release rates (HRRs), tunnel users situated within 30 meters from the fire source would not be able to escape safely.

Identification of Anthropogenic CO2 Using Triple Oxygen and Clumped Isotopes.

Quantification of contributions from various sources of CO2 is important for understanding the atmospheric CO2 budget. Considering the number and diversity of sources and sinks, the widely used proxies such as concentration and conventional isotopic compositions (δ13C and δ18O) are not always sufficient to fully constrain the CO2 budget. Additional constraints may help in understanding the mechanisms of CO2 production and consumption. The anomaly in triple oxygen isotopes or 17O excess (denoted by Δ17O) and molecules containing two rare isotopes, called clumped isotopes, are two recently developed tracers with potentials to independently constrain some important processes that regulate CO2 in the atmosphere. The clumped isotope for CO2, denoted by Δ47, is the excess of 13C16O18O over a random distribution of isotopes in a CO2 molecule. We measured the concentrations of δ13C, δ18O, Δ17O, and Δ47 in air CO2 samples collected from the Hsuehshan tunnel (length: 12.9 km), and applied linear and polynomial regressions to obtain the fossil fuel end-members for all these isotope proxies. The other end-members, the values of all these proxies for background air CO2, are either assumed or taken as the values obtained over the tunnel and ocean. The fossil fuel (anthropogenic) CO2 end-member values for δ13C, δ18O, Δ17O, and Δ47 are estimated using the two component mixing approach: the derived values are -26.76 ± 0.25‰, 24.57 ± 0.33‰, -0.219 ± 0.021‰, and 0.267 ± 0.036‰, respectively. These four major CO2 isotope tracers along with the concentration were used to estimate the anthropogenic contribution in the atmospheric CO2 in urban and suburban locations. We demonstrate that Δ17O and Δ47 have the potential to independently estimate anthropogenic contribution, and the advantages of these two over the conventional isotope proxies are discussed.

Model Analysis of Smoke Control in Long Tunnel: Findings from Hsueh-Shan Tunnel Accident in Taiwan

The common properties of risk in long tunnel fires are high temperature, extreme difficulty of evacuation, rescue urgency and obstacle to rescue operation. Therefore, a complete ventilation design is an indispensable safety measure. Hsueh-Shan Tunnel is the longest in Taiwan, the fifth longest in the world. On May 7, 2012, a serious tunnel fire caused two deaths and numerous victims suffered from smoke inhalation injury. Apart from this, there was smoking entering the cross-passages and shafts which were important for evacuation. In this research, the current ventilation system in Hsueh-Shan Tunnel was simulated with FDS (fire dynamics simulator) software, and the statistics of smoke, visibility and temperature profile were analyzed. The results of this research showed that, with the current ventilation system, the time was shorter and the distance was longer for the smoke spreading windward than in other models. Furthermore, the visibility of windward victims was more affected and the temperature above the fire source was higher than those in other systems. When the wind speed in tunnel is within 2.0~4.0 m/s, the condition for turning off the ventilation fan within 250 m upwind from the fire source can be prominently reduced to 50 m upwind from the fire source. This not only could avoid plume disturbance but also could be maintained. If victims' evacuation should be given the highest priority, it is recommended to straightly activate the maximum power of the fan. Language: en

Characteristics of trace metals in traffic-derived particles in Hsuehshan Tunnel, Taiwan: size distribution, potential source, and fingerprinting metal ratio

Abstract. Traffic emissions are a significant source of airborne particulate matter (PM) in ambient environments. These emissions contain an abundance of toxic metals and thus pose adverse effects on human health. Size-fractionated aerosol samples were collected from May to September 2013 by using micro-orifice uniform deposited impactors (MOUDIs). Sample collection was conducted simultaneously at the inlet and outlet sites of Hsuehshan Tunnel in northern Taiwan, which is the second-longest freeway tunnel (12.9 km) in Asia. This endeavor aims to characterize the chemical constituents and size distributions, as well as fingerprinting ratios of particulate metals emitted by vehicle fleets. A total of 36 metals in size-resolved aerosols were determined through inductively coupled plasma mass spectrometry. Three major groups – namely, tailpipe emissions (Zn, Pb, and V in fine mode), wear debris (Cu, Cd, Fe, Ga, Mn, Mo, Sb, and Sn), and resuspended dust (Ca, Mg, K, and Rb) – of airborne PM metals were categorized on the basis of the results of enrichment factor, correlation matrix, and principal component analysis. Size distributions of wear-originated metals resembled the pattern of crustal elements, which were predominated by super-micron particulates (PM1–10). By contrast, tailpipe exhaust elements such as Zn, Pb, and V were distributed mainly in submicron particles. By employing Cu as a tracer of wear abrasion, several inter-metal ratios – including Fe / Cu (14), Ba / Cu (1.05), Sb / Cu (0.16), Sn / Cu (0.10), and Ga / Cu (0.03) – served as fingerprints for wear debris. However, the data set collected in this work is useful for further studies on traffic emission inventory and human health effects of traffic-related PM.

Submicron particle characteristics of atmospheres in a long highway tunnel.

This study used a scanning mobility particle sizer (SMPS) to measure and categorize submicron atmospheric particles in the 14-737-nm size range for ambient and urban roadside air and for air in the Hsuehshan Tunnel (12.9 km), Taiwan. Principal component analysis, traffic flow, and particle size distributions were used to identify the emission characteristics of light-duty vehicles (LDV) with the SMPS data. In the Hsuehshan Tunnel, the particle size from the majority of emissions discharged by LDV is approximately 20-60 nm, and the maximum particle number can reach up to 2.5 × 10(5). In contrast, submicron particle size distribution for urban roadsides is mostly 14-200 nm, and the maximum particle number is approximately 4 × 10(4) with the particle number for most particle sizes being below 1,200. The submicron particle size distribution at the ambient air station was unimodal with a mode sizes at 30-50 nm with the maximum particle number of 3,000.

Numerical Analysis of Evacuation from Taiwan Hsuehshan Tunnel Fire

Taiwan is an island nation with numerous mountains and few plains. Consequently, the number of tunnel projects has gradually increased and tunnels are becoming longer. Because the number of large tunnels that exceed 1000 meters in length has increased, the effective escape and evacuation of people during a fire and the minimization of injury are crucial to fire protection engineers. For this study, an actual example of a fire that occurred in Hsuehshan Tunnel (12.9 kilometers and the longest tunnel in Southeast Asia) was used. A fire dynamics simulator (FDS) including numerical simulation software was applied to analyze this fire and the relevant information that was collected was compared and verified. The fire site simulation showed the escape and evacuation of people during the fire. Simulations of the original fire site and the possible escape time for people with various attributes were discussed to provide quantitative data and recommendations based on the analysis results, which can serve as a reference for fire protection engineering.

Risks of TBM Construction and Corresponding Measures

To get better understanding of TBM construction risk and avoid it, comprehensive analysis and summary of various factors to TBM construction risk were made in three aspects, such as geological and hydrological conditions, TBM equipment, construction team of TBM. Based on the detailed presentation of hydrological conditions, the corresponding demands of TBM equipment and construction team were put forward. Emphatically from the two aspects of bad geological and countermeasures, design of TBM equipment, corresponding strategies to avoid all kinds of risk were proposed. Using Taiwan Hsuehshan Tunnel as the class project, this paper introduced and analyzed the construction risk and of solutions to it, with the hoping of providing guidance for TBM design, risk analysis and evade of construction and other aspects.

Simulation of Bus Fires in the Hsuehshan Tunnel, Taiwan, Using the Software Fire Dynamics Simulator

Simulation of Bus Fires in the Hsuehshan Tunnel, Taiwan, Using the Software Fire Dynamics Simulator

The Effect of Developing a Tunnel across a Highway on the Water Quality in an Upstream Reservoir Watershed Area—A Case Study of the Hsuehshan Tunnel in Taiwan

Cities in Taiwan are so dependent on reservoir water that preservation of the upstream reservoir watershed has become a significant public concern. However, due to the high-density development of land, resulting in rapid urban expansion, the construction of tunnels and elevated highways across reservoirs to better utilize the surrounding land has become a global trend. Based on data from long-term observation of the reservoir, this study verifies the difference in water quality before and after the highway construction. The results indicate that the total phosphorus (TP) increased on average 14 μg/L to 36.5 μg/L per annum, and the water quality is expected to require 10 years to recover. During the highway development, the average TP was more than twice the normal level. During summer, the TP level increases 3.1-fold due to rainfall. As indicated by the results, the large-scale land development will harm the long-term preservation of the reservoir’s water quality, and therefore should be avoided.

地下水向雪山隧道的排泄对台湾北部水资源的影响

The Hsueh-Shan tunnel is the fifth longest road tunnel in the world. During the excavation, the tunnel encountered several events of groundwater inrush, causing serious delay of the construction. Data on groundwater discharge to the tunnel were gathered from the monitoring system and their spatial and temporal variations were analyzed. The results of the integrated analysis of groundwater discharge and local precipitation indicated that the discharge increased rapidly when the cumulative rainfall exceeded 85 mm. The groundwater level recession rate after a rainfall event was found to be independent of rainfall intensity. A hydrogeological conceptual model was developed to simulate the long-term groundwater discharge to the tunnel. Sensitivity analysis was first conducted to identify sensitive parameters, and then the calibration process was accomplished by the automated parameter estimation method. The calibrated model was then used to evaluate the potential impact of tunnel excavation on the Feitsui reservoir; the average percentage loss of inflow to the Feitsui reservoir from 2006 to 2010 is estimated to be 1.74 %. The developed model can provide a tool for evaluating the regional hydrogeologic setting and the influence of tunnel construction on water resources.