A biomass briquette, which are highly densified biomass raw material and a type of renewable energy are expected to suppress the global warming. To effectively utilize biomass briquettes, it is important to know their combustion characteristics. It is reported that the flaming combustion duration is proportional to the product of briquette density and inverse square of specific surface area, and the char combustion duration is proportional to the product of fixed carbon density and inverse square of specific surface area, with a single cylindrical briquette of which aspect ratio is 2 in the constant hot air condition. Each proportional constant, Cf for flaming combustion and Cc for char combustion, can be considered as the indicators of the biomass briquette combustion rate, may change with the briquette's aspect ratio. Previous studies have reported that the flaming combustion duration and the char combustion duration are both affected by the aspect ratio. However, the extent and tendency of each influence and the reason why such influences occur have not been clarified. In the present study, combustion experiments of a single cylindrical briquette were conducted to clarify the trends and extent of the influence of the briquette aspect ratio on both combustion durations.
This study aims to clarify the effect of inner-diameter differences on the delayed onset of flashback, where the H2/air premixed flame, after being trapped near a wire mesh, passes through the mesh and flashback occurs with a time delay. In the experiment, as the inner diameter decreased, the flashback range decreased, and the time from the flame trap near the mesh to the flashback increased. This is because the heat loss to the tube wall increased as the inner diameter decreased. On the other hand, the mesh temperature just before the start of flashback decreased as the inner diameter decreased. This is because heat recirculation is greater as the inner diameter decreases due to the longer time from the flame trap near the mesh to the flashback and the low flow rate of the mixture. Additionally, flame-front position was calculated using image analysis, and flame propagation behavior was examined in terms of the apparent flame speed. As a result, the findings of this study clarified the effects of heat loss and heat recirculation differences due to variations in the inner diameter on flame behavior near the mesh. This study provides knowledge on flame behavior near the flame arrester under flow conditions and contributes to the safe use of hydrogen.
Radiant tube heaters are widely used in industrial furnaces, and improving their overall thermal efficiency, including heat recovery, is essential for energy conservation and decarbonization. In this study, a computational fluid dynamics (CFD) model of a W-type gas-fired radiant tube equipped with a recuperator was developed, in which combustion, heat transfer, and air preheating by exhaust gas are explicitly and self-consistently coupled. To the best of our knowledge, a fully coupled simulation framework that dynamically integrates recuperator heat recovery with flow, reaction, and heat transfer in a radiant tube system has not yet been quantitatively demonstrated in the literature. The validity of the proposed model was confirmed through comparison with experimental data obtained from a radiant tube heating test furnace, showing good agreement in terms of radiant tube surface temperature distribution and heat balance under various operating conditions. The numerical analysis clarified that the effect of the recuperator on heat transfer is primarily localized in the upstream region of the radiant tube, where the heat flux from the combustion gas to the tube wall is significantly enhanced. In contrast, the increase in tube surface temperature observed in downstream sections is mainly attributed to enhanced radiative heat exchange between adjacent tube segments rather than direct convective heat transfer. Furthermore, air preheating by the recuperator maintains a higher velocity of the secondary air jet relative to the surrounding mixture, sustaining entrainment and promoting mixing and combustion reactions in the upstream region. As a result, the sensible heat of the exhaust gas is effectively reduced, and the contribution of the recuperator to thermal efficiency improvement becomes more pronounced with increasing fuel flow rate. These findings provide quantitative insight into the mechanisms by which recuperative air preheating influences flow, reaction, and heat-transfer characteristics, and demonstrate the importance of accounting for heat recovery performance in the design and evaluation of high-efficiency radiant tube burner systems.
The calcination behavior of powdered limestone particles in a pilot-scale combustion furnace was investigated numerically. Gas-temperature and outlet particle-size measurements for the methane-firing condition at an excess-air ratio of 1.5 were compared with the numerical predictions. No experimental calcination data directly comparable with the calculated outlet-averaged CaO mass fraction were available for the simulated cases. Therefore, the calculated CaO mass fractions are presented as numerical predictions rather than quantitatively validated values. Large-eddy simulation (LES) coupled with a flamelet/progress-variable (FPV) combustion model was employed to describe turbulent combustion, while limestone particles were tracked using a Lagrangian particle approach. A particle size evolution model associated with the progress of calcination was introduced to account for particle size reduction during decomposition. At the available measurement locations for the methane firing case, the predicted gas temperature levels were comparable to the experimental values. The particle size evolution model, calibrated using the measured outlet distribution for this condition, reproduced the main features of the distribution under the calibration condition. The effects of methane combustion and methane-ammonia co-firing were also investigated under identical thermal input conditions. Although the co-firing case exhibited different flow and temperature fields from the methane-only case, the calculated outlet-averaged CaO mass fractions were comparable. This behavior was associated with the combined effects of the flow field, the axial location of particle heating, and the particle residence time. The influence of the excess-air ratio was additionally examined. Increasing the excess-air ratio shifted the high-temperature region downstream and reduced the particle residence time. Consequently, the highest calculated calcination degree was obtained under the stoichiometric condition owing to the farther-upstream high-temperature region and the longer particle residence time.
Thermal therapy, including hyperthermia and thermal ablation, is minimally invasive and associated with minimal side effects. To ensure effective and safe thermal therapy, the temperature distributions inside the patient's body must be accurately measured. In clinical practice, thermal sensors are often adopted but are invasive and limited to point measurements. Alternatively, noninvasive temperature monitoring can be achieved through magnetic resonance imaging but presents several disadvantages, including expensive implementation and numerous contraindications. To address these drawbacks, this study investigates the feasibility of estimating two-dimensional temperature distributions from ultrasound images using a generative adversarial network (GAN). Because the speed of sound in the propagation medium depends on temperature, the increasing temperature induced by thermal therapy causes apparent scatterer displacements between ultrasound images acquired before and after heating. The proposed GAN-based method was trained using paired ultrasound images and corresponding ground-truth temperature distribution images generated by extending one-dimensional theoretical temperature profiles. Experiments were conducted using an agar phantom under first-(Dirichlet) and second-kind (Neumann) boundary conditions to evaluate whether the proposed GAN-based method could learn the mapping between paired ultrasound images and the corresponding model-generated ground-truth temperature distributions. The mean absolute error (MAE) between the theoretical model and estimated temperature distributions was evaluated within the effective thermal region, where the theoretical temperature increase was at least 10% of its maximum value at each heating time. The MAE within this region was 0.80 +/- 0.78 degrees C under the first-kind condition and 0.88 +/- 0.62 degrees C under the second-kind condition. Although larger local errors were observed in regions with steep temperature gradients, these results demonstrate the feasibility of the proposed GAN-based method as a proof-of-concept for noninvasive temperature estimation from ultrasound B-mode images using model-generated ground-truth temperature distributions.
