No abstract

In this paper, we proposed an innovative Bayesian optimization (BO) coupled with deep learning for rapid airfoil shape optimization to maximize aerodynamic performance of airfoils. The proposed aerodynamic coefficient prediction model (ACPM) consists of a convolutional path and a fully connected path, which enables the reconstruction of the end-to-end mapping between the Hicks–Henne (H–H) parameterized geometry and the aerodynamic coefficients of an airfoil. The computational fluid dynamics (CFD) model is first validated with the data in the literature, and the numerically simulated lift and drag coefficients were set as the ground truth to guide the model training and validate the network model based ACPM. The average accuracy of lift and drag coefficient predictions are both about 99%, and the determination coefficient R2 are more than 0.9970 and 0.9539, respectively. Coupled with the proposed ACPM, instead of the conventional expensive CFD simulator, the Bayesian method improved the ratio of lift and drag coefficients by more than 43%, where the optimized shape parameters of the airfoil coincide well with the results by the CFD. Furthermore, the whole optimization time is less than 2 min, two orders faster than the traditional BO-CFD framework. The obtained results demonstrate the great potential of the BO-ACPM framework in fast and accurate airfoil shape optimization and design.

Catalyst durability is one of the critical challenges for the commercialization of proton exchange membrane fuel cells. In this study, a one-dimensional model of a fuel cell cathode catalytic layer (CCL) is proposed to investigate the structural evolution, electrochemical surface area, Pt and Co loss of Pt-Co shell-core structured catalysts, and then the aging mechanism of catalyst is elaborated by simulation. The model considers three main processes: oxidation and redeposition of Pt on the Pt shell, crossover H2 through the membrane to reduce Pt2+ near the CCL/membrane interface, and leaching and dissolution of Co. The results show that the severe dissolution of catalyst particles near the CCL/membrane interface not only leads to a large loss of Pt and Co, but also causes the catalyst to age unevenly along the CCL thickness direction. In addition, both the increase in temperature and the decrease in the average particle size accelerate the catalyst aging.

Three-dimensional simulations were performed for proton exchange membrane fuel cell (PEMFC) with thin catalyst-coated membrane (CCM) regarding liquid water cooling design. The studied PEMFC follows a counter-flow pattern for the H 2 and air stream, which is commonly adopted in today's automotive PEMFCs. For the thermal modeling of the liquid water, conjugate heat transfer model is used. The cooling flow inlet temperature between 60 and 75 C, direction, flow rate between 0.08 and 0.32 L s À1 m À2 as well as the cooling channel number are investigated, specifically. It is found that the cooling inlet temperature directly determines the working temperature of PEMFC under the same cooling flow rate. It means that increasing the cooling inlet temperature can lift the PEMFC operating temperature. The co-direction for the liquid flow and the air stream is found to be better for PEMFC as it can suppress the liquid water formed near cathode outlet. It is then pointed out that the cooling flow rate would determine the along-channel temperature non-uniformity in PEMFC and moderate flow rate is preferred. Reducing the number of the cooling channels while assigning higher flow rate for each channel will slightly lift the PEMFC temperature overall, but this strategy will result in more pumping power loss.

In real electric vehicles, the arrangement of liquid-cooled plates not only influences the thermal performance of the battery pack but also relates to the energy consumption of the BTMS and the compactness of the whole battery pack. In this study, design A, design B, design C, and design D, a total of four different arrangement designs of battery thermal management based on liquid-cooled plates with microchannels, are proposed for a 35 V battery pack composed of 12 LiFePO4 pouch battery cells connected in series, and the corresponding three-dimensional electrical-thermal-fluid model is established for numerical study. The cooling effects of the four designs are discussed and compared in terms of discharge rate, contact thermal resistance, and external short circuit. For design D, cold plates are placed in front of each battery cell. The results show that design D achieves the best cooling effect with the lowest power consumption compared to the other three designs under 0.5C, 1.0C, and 2.0C discharge rate. Its maximum temperature is about 30°C, and maximum temperature difference is under 5°C. The reduction in contact thermal resistance has different effects and magnitudes for different designs with different cold plate arrangements, but the overall effect is small. In the extreme condition of external short circuit, for design D, increasing the mass flow rate can reduce the maximum temperature of design D from 76.6°C by 27.5% to 55.5°C and the temperature difference from 35.0°C by 23.4% to 26.8°C. Selecting the proper coolant flow rate can keep the maximum temperature and temperature gradient on the battery pack of design D within tolerable level, and increasing the flow rate helps to enhance the cooling effect. For the other three designs, the maximum temperatures and temperature gradients exceeded 90°C and 40°C under the external short circuit condition, and increasing the flow rate has very little effect on the performance enhancement.