Dynamic prediction of air conditions within a car cabin is significant to technology design and usage for thermal comfort, air quality, and energy consumption of vehicles. In this paper, theoretical models are completely derived from mass and energy balances of moist air and carbon dioxide from an outdoor environment via vehicle envelope to car cabin. However, it is demanding to determine genuine modeling hypothesis and/or physical properties of car cabins for accurate prediction. Regardless of those requirements, artificial neural networks can be applied as universal models, which are derived from numbers of input/output data from real dynamic systems. Without any experimental efforts, the input/output data is generated from the theoretical models under various conditions. A few input/output data from experiments are combined for making prediction close to actual behaviors. With bilateral analyses, the artificial neural networks are trained effectively to simulate dynamic behaviors of air, moisture, and carbon dioxide within a car cabin. This viability of the proposed methodology is confirmed by that the averaged coefficients of determination to the perfect prediction under parking and driving conditions of a sedan car are R²=0.9394, and R²=0.9314, respectively. The numbers of experimental data for training are 3.5% of total numbers of training data.

Oxygen combustion of CH₄-CO₂ mixture in a small-scale counterflow burner was studied. A diffusion flame was stabilized in the gap between opposed ports ejecting fuel and oxygen gases, and its thickness which depended on the gap distance between burner ports, inner diameter of burner tubes, flow rate of gases and fuel gas component was measured. The gas flow rates were varied such that the apparent equivalence ratio became constant at unity, while the gap distance was varied in the range from 1 mm to less. It was shown that the flame thickness decreased monotonically as the gap distance decreased and that the diffusion flame became thinner when the methane concentration in fuel gas became lower. Increased flame thickness was observed at large gas flow rate, and a diffusion flame was found in smaller gap distance at larger inner diameter. From these data, the relation between the flame thickness and the flame stretch rate was summarized, showing that the flame thickness decreased as the flame stretch became stronger, i.e. the thickness varied inversely with the square root of stretch rate. To elucidate the dependence of the flame thickness on the flame stretch rate, the flame thickness normalized by the average velocity of oxygen gas and the inner diameter was introduced. It was confirmed that the normalized flame thickness depended only on the flame stretch rate.

A thermoacoustic heat pump heater is a device having the ability to produce heat from acoustic power. The heat pump investigated was composed of an acoustic driver, a branched tube, and a looped tube. A porous component composed of many stacked steel screen meshes, called regenerator, was installed inside the looped tube. When an acoustic wave is supplied to the looped tube, filled with nitrogen gas at 0.5 MPa, heat pumping spontaneously occurs inside the regenerator. A numerical code was developed and tested. Then, the coefficient of performance was calculated for different temperatures by optimizing the regenerator flow channel radius and position inside the looped tube. It was found that the temperature affects considerably the regenerator optimum parameters.

Direct numerical simulations of thermal turbulent boundary layer flows over a wavy wall surface are performed to investigate the effect of a wavelength on drag coefficients and heat transfer performance. The Reynolds number based on an inlet boundary layer thickness is set to be 2820 and the Prandtl number is set to be Pr = 0.71 and 2.0. The wavy wall surface is homogeneous in the spanwise direction and the wave amplitude is fixed at 2a⁺ = 20. The six wavelength cases of λ/2a = 7.5 ～ 45 are examined. As the wavelength decreases, the skin-friction drag decreases and the pressure drag and heat transfer increase. The total drag peaks at λ/2a = 12.5 and the flow separation occurs at λ/2a < 15. In the separation region, the backward flow transfers the heat and results in a negative correlation coefficient between the velocity and temperature of R (u^′_tT^′ ) at the bottom of the wavy wall. Spindle-shaped spots of the Nusselt number are also observed on the upslope of the wavy wall.

Electrical vehicles equipped with lithium-ion batteries (LiBs) have been increasing in popularity on the market. LiBs have high energy density and high electric current; however, their lifetimes and performance are known to be strongly influenced by temperature rise due to heat generation, and thermal runaway may occur when the battery temperature exceeds 80°C. Hence, the development of LiB thermal-management technology is essential. In this study, an A4-sized LiB was short circuited in a prototype cooling system with phase-change material (PCM) and heat pipes (HPs), and the performance of the cooling system was evaluated. To compare the cooling performances, four experimental conditions were adopted: a combination of PCM and HP; PCM only; HP only; and not using the cooling system. In addition, a simulation was conducted under the experimental conditions using a scale model of the cooling system. Thus, we confirmed that the temperature increase of the LiB, especially up to 80°C, was extended by the effects of PCM. The combination of PCM and HP suppressed the temperature of LiB to be about 80°C.

In polymer electrolyte fuel cells (PEFCs), a gas diffusion layer (GDL) is a critical component to prevent flooding and to improve the cell efficiency under high current density operation. To advance the experimental method for evaluation of liquid water transport in GDL, this study proposes a method, termed scale model experiments. In this method, enlarged GDL structures are reproduced by a 3D printer, and simulated water behavior is observed with similarity conditions satisfying the flow in the GDL. The lattice Boltzmann simulation is applied to the enlarged model experiments and identifies dominant forces in the water dynamics. First, simulations are conducted to compare flow patterns at two different scales of GDL, the actual scale and 313 times enlarged structures, with combinations of two immiscible fluids. The results suggest that fluid behavior can be considered similar at the different scales, when the Capillary number is low enough for the flow to be dominated by capillary forces, at Ca = 3.0×10^-3, with the additional condition of negligible buoyancy. Next, experiments with two types of 313 times enlarged GDL structures are conducted with silicone oil and water of similar densities, and the flows are compared to the simulation results. These suggest that the water transport in the GDL can be successfully reproduced by the enlarged model experiments. As similarity conditions, the Weber number must be kept below the order of 10^-1 to suppress inertial forces as well as the Capillary number must be below the order of 10^-3 for smaller viscous forces. Careful attention must be paid to the viscous forces of the two fluids in a relatively-uniform structure GDL. Finally, an experiment of water transport in the GDL with a channel is demonstrated, showing the effectiveness of the proposed experimental approach for designing GDL structures tolerant to flooding.

In order to meet the needs of designing for high performance secondary air system in the turbine, a new turbine bite engineering rim seal of type-A and the related test rig were designed and set up. The numerical investigation and the experimental verification of the leakage characteristic of rim seal were carried out based on k - ω SST turbulence model. The key feature of the engineering rim seal of type-A, a bite type, was extracted, hence the other three kinds of rim seal named type-B, type-C and type-D were proposed. The sealing mechanism of bite rim seal was revealed from two aspects of the leakage characteristic and the seal effectiveness. The results show that: the numerical results of the k-ω SST turbulence model agree well with the experimental ones, and the maximum error is less than 4.2% in the range of this paper. The leakage of turbine bite rim seal of type-C and type-D are less than that of conventional rim seal (type-B). Compared with type-C, the radial inward shoulder structure of type-D reduces the leakage in the cavity, however, it enhances the pumping effect of the rotor. The average sealing efficiency of type-C and type-D are increased by 8.9% and 8.82% compared with that of conventional rim seal of type-B, respectively. The effects of entrainment and diffusion of the vortex in the sealing channel are one of the important reasons for the ingestion. There are two kinds of vortex structure named recirculation zone vortex (RZV) and rotationally induced vortex (RIV). The RZV will be rapidly replaced by the RIV with the rotation speed increasing at a low flow rate of the hot gas. However, the RZV will be relatively stable while the hot gas mass flow rate is increased.

The present work is concerned of numerical simulation of three dimensional laminar forced and mixed convection of two nanofluids; Al₂O₃-water and Cu-water flowing through a horizontal tube submitted to a constant and uniform heat flux. Based on single-phase approach, three dimensional conservation equations of mass, momentum and energy with the appropriate boundary conditions have been solved using finite volume method with the schemes of spatial and temporal discretization of second order precision and by using the SIMPLER algorithm with the Tri-Diagonal Matrix Algorithm (TDMA). At a fixed Reynolds number Re = 300 and Grashof number equal to 0 and 5×10⁵. The results show an increase in heat transfer ratio compared to pure water at several volume fractions for both alumina and copper based nanofluids. At a fixed volume fraction φ = 4%, the axial Nusselt number does not increase significantly in forced convection case. However, in mixed convection case the axial Nusselt number augments considerably especially with Cu-water nanofluid. On the other hand, secondary flow and axial velocity are slightly affected by nanoparticles volume fraction. It is proved in this study that nanofluids can also contributes to optimise pipes compactness, using 2% and 4% of alumina and copper respectively dispersed in water flowing through a pipe with given length gives higher axial Nusselt number ratio compared to pipes larger length but containing pure water.

The Intelligent Predetermined Control Method (IPCM) aims to predetermine the requirements of programmed changes in the test chamber, which differentiates from the gain-scheduling PID (GS-PID) control method in accordance with dry and wet bulb temperature errors. The IPCM consists of a scheduling of optimal control path, a determination of the heat and water vapor mass flow rates along the path, an estimation of heat loss from walls of chamber, and a compensation for unexpectedness with a fixed-gain PID controller. In order to verify the feasibility of the sinusoidal temperature test by using the IPCM, a cylindrical test chamber with independent flow loops was designed, and it equips a precise heating system and a variable refrigerant flow (VRF) cooling system with a variable-frequency drive (VFD) compressor. This paper uses process models with nonlinear gain to identify the models of the heating and cooling systems individually. An experiment was performed by using a 5-45 °C sine wave excitation. The results show that the temperature in the chamber closely tracks along the programmed sine wave excitation with less than ±0.25°C of errors in almost entire cycles, which indicates that the IPCM is able to advance the chamber testing into the acceleration test beyond the ordinary step and ramp excitation.

We propose a new DRG-based mechanism reduction method that considers heat release rate (HRR), transport of species, and reaction rate. In the original DRG method developed by T. Lu and C. K. Law, species importance is evaluated using only the reaction rate, and PSR is used for sampling. However, combustion phenomena are also affected by the transport of species and HRR, and sampling using PSR may be inadequate to produce a skeletal mechanism that can simulate a realistic flame. Therefore, in the proposed transport and heat DRG method, a 1D premixed flame is used for sampling, and species importance is evaluated relative to reaction rate, transport flux, and HRR. The proposed method was evaluated in ethylene/air and n-butane/air 1D premixed flame, and the laminar flame speed and flame structure were obtained using skeletal mechanisms created by three DRG methods, i.e., the original, transport, and transport and heat methods. In the case of the ethylene/air premixed flame, the transport and heat DRG method produced a smaller skeletal mechanism that well-reproduced the result of the detailed mechanism. In the case of n-butane/air premixed flame, smaller skeletal mechanisms were obtained in the region of relative error on laminar flame speed of 0.1–1.0% with the transport and heat DRG method.

Forced ignition is important for the scramjet engines in supersonic flights. When compared to other forced ignition devices, burned-gas torch igniters are considered suitable for scramjet combustors because it is easy to control the input energy of burned-gas torch igniters. In the conventional burned-gas injection methods, the injection gas total temperature is estimated by assuming a discharge coefficient in the range of 0.90-0.95. However, the assumption has not been validated and it may cause misestimations when analyzing combustion in scramjet engine in supersonic flight. Therefore, a validated temperature estimation method is needed. In this paper, a new total temperature estimation method was proposed and investigated experimentally and numerically. Additionally, a novel hydrogen/air burned-gas torch igniter, which can control the injection gas total temperature by controlling the overall equivalence ratio, was also developed. The experimental results showed that the newly-developed hydrogen/air burned-gas torch igniter can successfully control the injection gas total temperature by changing the overall equivalence ratio in the range of 1.0-10.0. Also, the chemical equilibrium assumption for estimating injection gas compositions was validated by O₂ mole fraction measurement. The numerical results showed that the estimated discharge coefficient results in a maximum underestimation of 7.02% due to wall heat loss, which means that the maximum uncertainty of the estimated total temperature is 14.7%. This, in turn, indicates that the influence of wall heat loss on the discharge coefficient should be considered to estimate the injection gas total temperature more accurately.

A series of numerical simulations are performed to study the internal crossflow effect on single-row film cooling performances on a turbine blade suction surface, under the representative film-cooled engine-simulated conditions. The cylindrical and fan-shaped holes are considered, both having the same length-to-diameter of 3. In the current simulations, the blowing ratio (Br ) is selected as 1, 2 and 3 respectively. The velocity ratio of the internal crossflow to ejection jet (Vr ) is selected as 0, 1 and 2, respectively. The results show that the helical flow feature is dominant inside film hole with the presence of internal crossflow and the ejection jet is pushed toward one side in accordance to the internal flow direction. For the fan-shaped hole, the effect of internal crossflow on mutual interaction between ejection jet and primary flow downstream the film cooling hole is relatively weaker in compared to the cylindrical hole. In general, the velocity ratio of Vr=1.0 has a little influence on the film-hole discharge coefficient. However, under a high velocity ratio of Vr=2.0, approximately 15%~20% reduction of discharge coefficient is produced for the cylindrical hole, and 20%~27% reduction for the fan-shaped hole. In the viewing of laterally-averaged adiabatic film cooling effectiveness, the impact of internal crossflow effect is more profound for the cylindrical hole. In general, a moderate velocity ratio (Vr=1.0) plays a positive role on film cooling improvement but a high velocity ratio (Vr=2.0) is confirmed to reduce the film cooling effectiveness.

Development of a supersonic nozzle with a rectangular cross section for a scramjet model combustor was performed and verified experimentally and numerically. The newly-proposed combined design methodology is easier to use than the other design methodologies when the nozzle exit geometry is the constraint condition because it can get information about the displacement thickness before determining the supersonic nozzle contour. The velocity profile of the numerical simulation and that of LDV measurements are quantitatively in good agreement, which meant that the numerical simulations performed in the present study could quantitatively evaluate the flowfield in the supersonic nozzle and the isolator developed in the present study. The designed supersonic nozzle showed good uniformity of the Mach number and stream angle in the mainstream region at the nozzle exit. Also, the average Mach number in the mainstream region at the nozzle exit plane showed very good agreement with the design Mach number. The influence of airstream total temperature on the Mach number distribution and stream angle distribution in the mainstream region was investigated numerically and was acceptably small, which means that the supersonic nozzle designed in this study can create approximately equivalent supersonic flow in the range of airstream total temperature 300 K - 800 K. Additionally, The mainstream region was found to have a trapezoidal shape with the upper side as the short side and the lower side as the long side. This trapezoidal shape is considered to be caused by the pressure gradient along the side wall, indicating that it is difficult to prevent the mainstream region from growing to trapezoidal shape in case of two dimensional supersonic nozzle with a single contour wall.

Physical Instability of the MPCM suspensions tends to suppress suspensions’ heat transportation in heat exchangers and thermal energy storage tanks. The aim of current work is to determine the optimal surfactant and its content, pH-value, and density of the carrier fluid by experimental investigation. In current work, the near infrared transmitting and backscattering fluxes were measured using a universal stability analyzer (TURBISCAN LAb). Such two fluxes characterize the physical stability. It was found, the sodium dodecyl sulfate (SDS) was the better surfactant among the selected surfactants. Experimental data indicates that the optimum SDS mass concentration ratio was 0.2wt.%. The best pH-value was 8 among the selected values. It was also found the MPCM suspension was the most stable at the carrier fluid density of 0.941 g•cm-3. The density difference between the dispersion phase and carrier fluid is the most significant parameter. A near zero density difference, combining the appropriate setup of the other parameters involving surfactant type, its concentration, pH-value, can improve the physical stability of MPCM suspensions significantly, enabling their applications in heat transfer and thermal energy storage.

Two dimensional unsteady calculations of reactive flows were performed in large domain to investigate the unstable behaviors of cellular premixed flames at low Lewis numbers based on the diffusive-thermal (D-T) model and compressible Navier-Stokes (N-S) equations including one-step irreversible chemical reaction. The relations between the growth rate and wave number, i.e. the dispersion relations, were obtained. The growth rates obtained by the compressible N-S equations were large and the unstable ranges were wide compared with those obtained by the D-T model equations. This was because both hydrodynamic and diffusive-thermal effects were taken into account in the compressible N-S equations. A disturbance with the maximum growth rate was superimposed on a stationary planar flame to generate the cellular flame. The cellular flame formed, and the overshoot of temperature evolved. When the length of computational domain increased, the number of small cells separated from large cells of the cellular flame increased drastically. As the results, flame surface area and average burning velocity increased. The stronger unstable behaviors and the larger average burning velocities were observed especially in the numerical results based on the compressible N-S equations. In addition, the fractal dimension obtained by the compressible N-S equations was larger than that by the D-T model equations. Moreover, we confirmed that the radiative heat loss promoted the instability of premixed flames at low Lewis numbers.

To better understand the performance of gaseous film cooling near the injector region in the LOX/GH₂ thrust chambers, a methodology is employed to simulate the coupled heat transfer of the film cooling in thrust chambers with regenerative cooling. The conjugated flow and heat transfer behaviors of film coolant, hot gas, cooling channels and regenerative coolant are numerically investigated. A three-dimensional non-adiabatic flamelet model using real gas equation of state is developed to solve the combustion and validated against the experimental data. Film cooling performance is predicted for the conditions with different geometrical parameters and mass flow rates of the film coolant. The result shows that the reverse flow zones are formed and developed in the region near the injectors The occurrence of those zones is responsible for the significant reduction of the hot-gas-side wall temperature near the head plate of the injector. The coolant mass flow rate has a great influence on the film cooling performance due to the variety of coolant momentum in the exit of orifices and vorticity in the near-injector region. An optimum mass flow rate for maximizing the averaged effectiveness exists for a given film orifice geometric configuration. The high averaged effectiveness and the uniform flux distribution of hot-gas-side wall are observed at small orifices spacing. The film cooling effectiveness in the orifice exit region is obviously enhanced when the diameter of the orifice increased in the front part of the combustion chamber. The results would be useful for the analysis and optimization design of the straight cylindrical coolant orifices in the LOX/GH₂ thrust chamber.