Numerical performance of CO2 accumulation and droplet dispersion from a cough inside a hospital lift under different ventilation strategies.

The impact of mechanical ventilation on airborne diseases is not completely known. The recent pandemic of COVID-19 clearly showed that additional investigations are necessary. The use of computational tools is an advantage that needs to be included in the study of designing safe places. The current study focused on a hospital lift where two subjects were included: a healthy passenger and an infected one. The elevator was modelled with a fan placed on the middle of the ceiling and racks for supplying air at the bottom of the lateral wall. Three ventilation strategies were evaluated: a without ventilation case, an upwards-blowing exhausting fan case and a downwards-blowing fan case. Five seconds after the elevator journey began, the infected person coughed. For the risk assessment, the CO2 concentration, droplet removal performance and dispersion were examined and compared among the three cases. The results revealed some discrepancies in the selection of an optimal ventilation strategy. Depending on the evaluated parameter, downward-ventilation fan or no ventilation strategy could be the most appropriate approach.

Cooktop Sensing Based on a YOLO Object Detection Algorithm.

Deep Learning (DL) has provided a significant breakthrough in many areas of research and industry. The development of Convolutional Neural Networks (CNNs) has enabled the improvement of computer vision-based techniques, making the information gathered from cameras more useful. For this reason, recently, studies have been carried out on the use of image-based DL in some areas of people's daily life. In this paper, an object detection-based algorithm is proposed to modify and improve the user experience in relation to the use of cooking appliances. The algorithm can sense common kitchen objects and identify interesting situations for users. Some of these situations are the detection of utensils on lit hobs, recognition of boiling, smoking and oil in kitchenware, and determination of good cookware size adjustment, among others. In addition, the authors have achieved sensor fusion by using a cooker hob with Bluetooth connectivity, so it is possible to automatically interact with it via an external device such as a computer or a mobile phone. Our main contribution focuses on supporting people when they are cooking, controlling heaters, or alerting them with different types of alarms. To the best of our knowledge, this is the first time a YOLO algorithm has been used to control the cooktop by means of visual sensorization. Moreover, this research paper provides a comparison of the detection performance among different YOLO networks. Additionally, a dataset of more than 7500 images has been generated and multiple data augmentation techniques have been compared. The results show that YOLOv5s can successfully detect common kitchen objects with high accuracy and fast speed, and it can be employed for realistic cooking environment applications. Finally, multiple examples of the identification of interesting situations and how we act on the cooktop are presented.

Numerical modeling of a sneeze, a cough and a continuum speech inside a hospital lift.

The global COVID-19 and its variants put us on notice of the importance of studying the spread of respiratory diseases. The most common means of propagation was the emission of droplets due to different respiration activities. This study modeled by computational fluid dynamics (CFD) techniques a high risk scenario like a hospital elevator. The cabin was provided with an extraction fan and a rack for air renewal. Inside, a sneeze, a cough and a continuum speech were simulated. Inside the lift, two occupants were introduced to observe the risk of propagation of emitted droplets and the impact in diseases spreading risk. The fan effectivity over the droplets ejection was analyzed, as well as environmental condition of a clinical setting. For this purpose the amount of droplets inside were counted during whole time of simulations. The effect of the fan was concluded as able to eject the 60% of small droplets, but also a high performance in spreading particles inside. Among the three cases, the riskiest scenario was the continuum speech due to the saturation of droplets in airborne.

CNN-based flow control device modelling on aerodynamic airfoils.

Wind energy has become an important source of electricity generation, with the aim of achieving a cleaner and more sustainable energy model. However, wind turbine performance improvement is required to compete with conventional energy resources. To achieve this improvement, flow control devices are implemented on airfoils. Computational fluid dynamics (CFD) simulations are the most popular method for analyzing this kind of devices, but in recent years, with the growth of Artificial Intelligence, predicting flow characteristics using neural networks is becoming increasingly popular. In this work, 158 different CFD simulations of a DU91W(2)250 airfoil are conducted, with two different flow control devices, rotating microtabs and Gurney flaps, added on its Trailing Edge (TE). These flow control devices are implemented by using the cell-set meshing technique. These simulations are used to train and test a Convolutional Neural Network (CNN) for velocity and pressure field prediction and another CNN for aerodynamic coefficient prediction. The results show that the proposed CNN for field prediction is able to accurately predict the main characteristics of the flow around the flow control device, showing very slight errors. Regarding the aerodynamic coefficients, the proposed CNN is also capable to predict them reliably, being able to properly predict both the trend and the values. In comparison with CFD simulations, the use of the CNNs reduces the computational time in four orders of magnitude.

Energy and thermal modelling of an office building to develop an artificial neural networks model.

Nowadays everyone should be aware of the importance of reducing CO2 emissions which produce the greenhouse effect. In the field of construction, several options are proposed to reach nearly-Zero Energy Building (nZEB) standards. Obviously, before undertaking a modification in any part of a building focused on improving the energy performance, it is generally better to carry out simulations to evaluate its effectiveness. Using Artificial Neural Networks (ANNs) allows a digital twin of the building to be obtained for specific characteristics without using very expensive software. This can simulate the effect of a single or combined intervention on a particular floor or an event on the remaining floors. In this paper, an example has been developed based on ANN. The results show a reasonable correlation between the real data of the Operative Temperature with the Energy Consumption and their estimates obtained through an ANN model, trained using an hourly basis, on each of the floors of an office building. This model confirms it is possible to obtain simulations in existing public buildings with an acceptable degree of precision and without laborious modelling, which would make it easier to achieve the nZEB target, especially in existing public office buildings.

Computational characterization of the behavior of a saliva droplet in a social environment.

The conduct of respiratory droplets is the basis of the study to reduce the spread of a virus in society. The pandemic suffered in early 2020 due to COVID-19 shows the lack of research on the evaporation and fate of droplets exhaled in the environment. The current study, attempts to provide solution through computational fluid dynamics techniques based on a multiphase state with the help of Eulerian-Lagrangian techniques to the activity of respiratory droplets. A numerical study has shown how the behavior of droplets of pure water exhaled in the environment after a sneeze or cough have a dynamic equal to the experimental curve of Wells. The droplets of saliva have been introduced as a saline solution. Considering the mass transferred and the turbulence created, the results has showed that the ambient temperature and relative humidity are parameters that significantly affect the evaporation process, and therefore to the fate. Evaporation time tends to be of a higher value when the temperature affecting the environment is lower. With constant parameters of particle diameter and ambient temperature, an increase in relative humidity increases the evaporation time. A larger particle diameter is consequently transported at a greater distance, since the opposite force it affects is the weight. Finally, a neural network-based model is presented to predict particle evaporation time.

Natural Ventilation Characterization in a Classroom under Different Scenarios.

The COVID-19 pandemic has pointed to the need to increase our knowledge in fields related to human breathing. In the present study, temperature, relative humidity, carbon dioxide (CO2) concentration, and median particle size diameter measurements were taken into account. These parameters were analyzed in a computer classroom with 15 subjects during a normal 90-minute class; all the subjects wore surgical masks. For measurements, Arduino YUN, Arduino UNO, and APS-3321 devices were used. Natural ventilation efficiency was checked in two different ventilation scenarios: only windows open and windows and doors open. The results show how ventilation affects the temperature, CO2 concentration, and median particle diameter size parameters. By contrast, the relative humidity depends more on the outdoor meteorological conditions. Both ventilation scenarios tend to create the same room conditions in terms of temperature, humidity, CO2 concentration, and particle size. Additionally, the evolution of CO2 concentration as well as the particle size distribution along the time was studied. Finally, the particulate matter (PM2.5) was investigated together with particle concentration. Both parameters showed a similar trend during the time of the experiments.

Delamination Fracture Behavior of Unidirectional Carbon Reinforced Composites Applied to Wind Turbine Blades.

One of the materials that is used widely for wind turbine blade manufacturing are fiber-reinforced composites. Although glass fiber reinforcement is the most used in wind turbine blades, the use of carbon fiber allows larger blades to be manufactured due to their better mechanical characteristics. Some turbine manufacturers are using carbon fiber in the most critical parts of the blade design. The larger rotors are exposed to complex loading conditions in service. One of the most relevant structures on a wind turbine blade is the spar cap. It is usually manufactured by means of unidirectional laminates, and one of its major failures is the delamination. The determination of material features that influence delamination initiation and advance by appropriate testing is a fundamental topic for the study of composite delamination. The fracture behavior is studied across coupons of carbon fiber reinforcement epoxy laminates. Fifteen different test conditions have been analyzed. Fracture surfaces for different mode ratios have been explored using optical microscope and scanning electron microscope. Experimental results shown in the paper for critical fracture parameters agree with the theoretically expected values. Therefore, this experimental procedure is suitable for wind turbine blade material characterizing at the initial coupon-scale research level.

Piezoelectric Energy Harvesting Controlled with an IGBT H-Bridge and Bidirectional Buck-Boost for Low-Cost 4G Devices.

In this work, a semi-submersible piezoelectric energy harvester was used to provide power to a low-cost 4G Arduino shield. Initially, unsteady Reynolds averaged Navier-Stokes (URANS)-based simulations were conducted to investigate the dynamic forces under different conditions. An adaptive differential evolution (JADE) multivariable optimization algorithm was used for the power calculations. After JADE optimization, a communication cycle was designed. The shield works in two modes: communication and power saving. The power-saving mode is active for 285 s and the communication mode for 15 s. This cycle consumes a determinate amount of power, which requires a specific piezoelectric material and, in some situations, an extra power device, such as a battery or supercapacitor. The piezoelectric device is able to work at the maximum power point using a specific Insulated Gate Bipolar Transistor (IGBT) H-bridge controlled with a relay action. For the extra power supply, a bidirectional buck-boost converter was implemented to flow the energy in both directions. This electronic circuit was simulated to compare the extra power supply and the piezoelectric energy harvester behavior. Promising results were obtained in terms of power production and energy storage. We used 0.59, 0.67 and 1.69 W piezoelectric devices to provide the energy for the 4G shield and extra power supply device.

Temperature based maximum power point tracking for photovoltaic modules.

In this article authors propose a temperature based Maximum Power Point Tracking algorithm (MPPT). Authors show that there is an optimal current vs maximum power curve that depends on photovoltaic (PV) module temperature. Therefore, the maximum power point (MPP) can be achieved in very few commutation steps if the control forces the PV module to work in temperature dependent optimal curve. Authors shows how this PV module temperature based MPPT is stable and converges to MPP for each temperature. In order to proof its stability, authors propose a Lyapunov energy function. This Lyapunov energy function has positive values for all values except into MPP given the PV module temperature. This Lyapunov energy function has negative increment along each time step. Hence, the stability of temperature based MPPT can be demonstrated. The proposed MPPT algorithm proposes a current set point. This current set point is obtained with instantaneous PV module power and temperature dependent maximum power vs optimal current curve. Stability is analysed for different temperature levels. Optimal current vs maximum power curve has been modelled by a line. The lines' coefficients depend on PV module temperature. Proposed Lyapunov energy function is not symmetric about equilibrium or MPP because MPPT algorithm and PV module dynamic have no symmetric behaviour about this equilibrium point.

Oscillating U-Shaped Body for Underwater Piezoelectric Energy Harvester Power Optimization.

Vibration energy harvesting (VeH) techniques by means of intentionally designed mechanisms have been used in the last decade for frequency bandwidth improvement under excitation for adequately high-vibration amplitudes. Oil, gas, and water are vital resources that are usually transported by extensive pipe networks. Therefore, wireless self-powered sensors are a sustainable choice to monitor in-pipe system applications. The mechanism, which is intended for water pipes with diameters of 2-5 inches, contains a piezoelectric beam assembled to the oscillating body. A novel U-shaped geometry of an underwater energy harvester has been designed and implemented. Then, the results have been compared with the traditional circular cylinder shape. At first, a numerical study has been carried at Reynolds numbers Re = 3000, 6000, 9000, and 12,000 in order to capture as much as kinetic energy from the water flow. Consequently, unsteady Reynolds Averaged Navier-Stokes (URANS)-based simulations are carried out to investigate the dynamic forces under different conditions. In addition, an Adaptive Differential Evolution (JADE) multivariable optimization algorithm has been implemented for the optimal design of the harvester and the maximization of the power extracted from it. The results show that the U-shaped geometry can extract more power from the kinetic energy of the fluid than the traditional circular cylinder harvester under the same conditions.

Arm Orthosis/Prosthesis Movement Control Based on Surface EMG Signal Extraction.

This paper shows experimental results on electromyography (EMG)-based system control applied to motorized orthoses. Biceps and triceps EMG signals are captured through two biometrical sensors, which are then filtered and processed by an acquisition system. Finally an output/control signal is produced and sent to the actuators, which will then perform the actual movement, using algorithms based on autoregressive (AR) models and neural networks, among others. The research goal is to predict the desired movement of the lower arm through the analysis of EMG signals, so that the movement can be reproduced by an arm orthosis, powered by two linear actuators. In this experiment, best accuracy has achieved values up to 91%, using a fourth-order AR-model and 100ms block length.