U. Fernandez-Gamiz

This article explores the novel application of a trained artificial neural network (ANN) in the prediction of vanadium redox flow battery behaviour and compares its performance with that of a two-dimensional numerical model. The aim is to evaluate the capability of two ANNs, one for predicting the cell potential and one for the overpotential under various operating conditions. The two-dimensional model, previously validated with experimental data, was used to generate data to train and test the ANNs. The results show that the first ANN precisely predicts the cell voltage under different states of charge and current density conditions in both the charge and discharge modes. The second ANN, which is responsible for the overpotential calculation, can accurately predict the overpotential across the cell domains, with the lowest confidence near high-gradient areas such as the electrode membrane and domain boundaries. Furthermore, the computational time is substantially reduced, making ANNs a suitable option for the fast understanding and optimisation of VRFBs.

In this paper, we are thus motivated to define and introduce the extended fuzzy-valued convex functions that can take the singleton fuzzy values − ∞ ˜ and + ∞ ˜ at some points. Such functions can be characterized using the notions of effective domain and epigraph. In this way, we study important concepts such as fuzzy indicator function and fuzzy infimal convolution for extended fuzzy-valued functions. Finally, we introduce the concept of directional generalized derivative for extended above functions and its properties. Eventually, we give a practical example that will illustrate well the directional g -derivative for the extended fuzzy-valued convex function.

The impact of the COVID-19 pandemic has caused an increase in knowledge in certain fields such as human breathing and respiratory diseases. In the current study, enclosed-space characteristics were assessed due to the great debate generated because of the pandemic. Temperature, relative humidity, carbon dioxide (CO2) concentration, particulate matter (PM) concentration and cumulative particle concentration percentage per size fraction were measured in three buildings under the occurrence of activities with different intensity levels. For the measurements, Arduino UNO, Sensiron SCD30 and APS-3321 devices were used. Natural ventilation was controlled via windows and/or doors, and the influence of air conditioning was studied in one of these rooms, in which medium- to high-intensity activities were conducted. All experiments revealed a decrease in temperature associated with an increase in the relative humidity when CO2 decreased. As a consequence, drastic decreases occurred a few minutes after windows and doors were opened. In no-ventilation intervals, the observed linear increase changed into an asymptotic trend below the 1000 ppm limit recommended by Sinphonie guidelines. Additionally, the opposite behavior was observed between particles above and below a certain size. This experiment is of interest to further investigate critical ventilation rate versus volume of air per individual.