Search

This paper represents a systematic approach to develop a virtual hydrogen sensor for predicting both the incidence and the extent of hydrogen starvation in high-temperature solid oxide fuel cells (SOFCs). Fuel starvation would occur in a fuel cell when the fuel is consumed at a faster rate than is fed to the cell. A previously developed and validated pseudo-2D numerical model was used for finding the hydrogen distribution along the channels of a single cell of a high-temperature SOFC. Nine different input parameters of the model were changed and a dataset of nearly half a million operating points was generated using the fast and accurate pseudo-2D model. The dataset was randomly divided into the training set (70%) and the test set (30%). Four different binary classifiers including K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), Naive Bayes, and Logistic Regression were employed to determine if the cell operates in normal (0) or starved (1) at a given set of input parameters. It was found that KNN and ANN outperform the other methods with an F1-Score above 0.97 if the parameters are appropriately tuned. Another ANN was then trained using the starved data to estimate the percentage of the cell being starved with an accuracy of 97.5% based on the mean absolute error (MAE). Therefore, the proposed set of classifier-regressor can be successfully employed as a virtual hydrogen sensor for online tracking of hydrogen concentration along the cell.

Among various types of energy conversion systems, hydrogen-fueled solid oxide fuel cells (SOFCs) have been acknowledged as one of the most promising technologies, thanks to the high energy density and numerous environmental and technical benefits they offer. The current study aims to provide a comprehensive review on numerical software-based modeling of these systems. The work is motivated by the increasing demand in utilization of software packages for investigating the multi-physiochemical phenomena occurring within the cell. The available software packages are introduced with the different applications being outlined. Novelty of each work as well as the corresponding drawbacks are represented through a critical approach. In the end, innovative designs and models, existing gaps in the literature, careful considerations for developing the numerical models, and recommendations for achieving further improvements in the field are provided.

In this paper, a numerical method is developed to simulate the effect of an external magnetic field on convection heat transfer of a ferrofluid flow inside a rectangular duct. This model is two-dimensional, steady-state, laminar and incompressible. Simple finite volume method is used to couple and solve continuity, momentum and energy equations. In the first part of this paper, it is observed that ferrofluid particles movement along the magnetic field of a line dipole changes the streamline patterns and enhances heat transfer. In the second part of this paper, a sensitivity analysis is performed to investigate the effects of Reynolds number, magnetic field strength, location, and number of line dipoles on the convection heat transfer. It is observed that cooling rate is higher at larger Reynolds numbers while increasing magnetic field strength would not necessarily result in noticeable improvement in heat transfer. It is also noticed that magnetic field strength should be large enough to merge small vortexes and to create large circulation zones inside the duct. Our simulation also revealed that heat transfer could be augmented by adding line dipoles on high-temperature regions closer to the duct inlet.

In the current study, a single cell of a planar SOFC is firstly modeled in 3D using commercial SOFC module of ANSYS Fluent and the results are validated against the experimental investigations in the literature. Many researchers have used ANSYS Fluent for simulating solid oxide fuel cells. However, there is a huge gap in the literature on explaining the detailed procedure that should be followed in order to use this software effectively. A thorough step-by-step approach is presented to provide a deep insight into the software. Thereafter, a simplified quasi-2D method with infinitely shorter computational time is developed and the results are compared with the 3D model. It is found that the reduced model is capable of being utilized as an alternate method for both online diagnosis and designing active control strategies.

In this paper, a new approach is presented to find the optimum catalyst loading distribution along the flow field. The optimization is performed by integrating a computational fluid dynamic (CFD) model and genetic algorithm optimization method. The CFD model is two-dimensional, steady state and two-phase. Multiphase mixture model (M 2 ) is used to model two-phase transport in porous media of a Polymer Electrolyte Membrane Fuel Cell (PEMFC). Numerical domain includes channel, gas diffusion layer (GDL) and catalyst layer (CL) in the cathode side. In the next step, current density is assumed to be proportional with catalyst loading. Catalyst loading is considered as polynomial functions with unknown coefficients. Genetic algorithm optimization method is applied to find the unknown coefficients and as a result the optimum catalyst loading function along the flow field. The results indicate that catalyst loading distribution has a significant effect on the fuel cell performance and it is seen that in the optimum case, maximum PEMFC power density is increased by about 14%.

In this study, the effect of addition of copper oxide nanoparticles on condensing heat transfer coefficient of R600a refrigerant flowing in a flat tube condenser has been investigated experimentally. The test setup consists of a pump, condenser test, second condenser, evaporator, heaters, and flow meter. The validation of the study was done by comparing the obtained condensation heat transfer coefficients with different empirical correlations in the literature. Different fluids including pure R600a, R600a-oil with Polyester oil (POE) mass percentage of 1%, and three R600a-oil-nanoparticle mixtures with mass percentages of 0.5%, 1%, and 1.5% were studied experimentally. It was shown that adding nanoparticles will result in 4.1%, 8.11%, and 13.7% average increase in condensing heat transfer coefficient with respect to the R600a-oil mixture. The greatest amount of increase was reported for the weight fraction of 1.5%, where it was observed that the condensing heat transfer coefficient for the mixture passing through the flattened tube is averagely 109.3% higher than its corresponding value for the pure refrigerant flowing in the round tube with the same mass flux. It was also found that an increase in mass flux resulted in an increase the heat transfer coefficient at all vapor qualities.

In this paper, the feasibility of a solar absorption refrigeration system to be powered by a latent heat storage (LHS) unit is investigated for a representative building. A single effect absorption chiller, utilizing Li-Br and water as working fluids is thermodynamically simulated. Then, the simulation of the latent heat storage unit is performed by applying finite difference method and the results were validated by the researches in the literature. Then, the geometry of a phase change material (PCM) based LHS system was optimized using multi-objective Genetic Algorithm for simultaneously minimizing the charging time, and maximizing the discharging time. Since the paper considers conflicting objectives, a Pareto front is presented that can be used for obtaining the optimum geometry according to the environmental conditions and working hours of the absorption system. As an illustrative example, the designed heat storage system was shown to be able to drive the 72 kW generator of an absorption system, for at least 10 h of operation in the discharging mode with the absence of sunlight. Therefore, it is possible to run absorption chillers under low-load operation conditions using the solar energy if the appropriate storage unit, such as what is introduced here, is used.

In this study, an improved concept design is presented to increase the thermal efficiency of the Rankine cycle of a typical steam power plant by combining a solar chimney and a dry cooling tower. The sources of the wind energy generation, include: the rejected heat from condenser to the air entering dry cooling tower, solar radiation and the airlift pumping effect on the air flow created by the stack hot flue gas which is injected into the hybrid tower as a novel change. This research primarily focuses on the Shahid Rajaee 250 MW steam power plant to determine the velocity of generated flow at the turbine inlet; a numerical finite volume code was employed for a dry cooling tower having a base diameter and a chimney height of 250 and 200 m, respectively. Calculations have been iterated for different angles of chimney walls, slopes of collectors and the base ground to find their effects on the output power. A range of 360 kW to more than 4.4 MW power is captured by the wind turbine by changing the hybrid tower geometrical parameters. Obtained results reveal a maximum of 0.538% increase for the thermal efficiency of the fossil fuel power plant.