Perovskite solar cells (PSCs) continue to advance toward higher efficiencies, yet the geometrical design of functional layers remains a critical bottleneck for device optimization and manufacturability. This work establishes a hybrid physics-data framework that integrates three-dimensional finite-element modeling with machine-learningbased surrogate prediction to accelerate PSC thickness optimization. A full 3D COMSOL Multiphysics model was developed to resolve charge-transport behavior, spatial electric fields, and recombination profiles within TiO₂/MAPbI₃/Spiro-OMeTAD architectures. Systematic variations in electron transport layer (ETL), perovskite absorber, and hole transport layer (HTL) thicknesses reveal that device power conversion efficiency (PCE) is governed by a trade-off between optical absorption, interface recombination, and resistive losses. A multi-layer perceptron regressor was trained using simulation data and achieved strong predictive fidelity (R2 ≈ 0.98) with a mean absolute error below 0.3%. The resulting surrogate model rapidly identifies optimal structural configurations without requiring additional high-cost simulations, demonstrating a reduction of design time by more than an order of magnitude. The proposed workflow provides a transferable route toward digital-twin-driven photovoltaic design and offers practical guidance for high-performance PSC engineering with reduced material consumption and enhanced computational efficiency.

Electromyography (EMG)-based hand gesture classification is a developing core technology for designing intuitive and responsive human-computer interaction, notably for prosthetic control. EMG signals, which reflect muscle activity during contraction, offer a non-invasive and effective method for capturing user gestures. However, because of their natural variability, noise, and temporal richness pose significant hurdles to precise gesture recognition. In this paper, we investigate the use of causal convolutional layers, which are suitable for sequential data, to improve hand gesture recognition from raw EMG signals. We propose a deep neural network which bases on temporal convolutions and integrates residual connections and contextual attention in an end to end hand gesture recognition system. Furthermore, we apply multiple data augmentation techniques to mitigate intra-subject variability and enhance model generalization. Our approach is evaluated on the benchmark NinaProDB1 dataset. The proposed model show impressive classification performance with an average accuracy of 95.31% and where the majority of the gestures from various subjects were accurately recognized. These results demonstrate the effectiveness of causal convolutions and attention mechanisms for robust EMG-based gesture recognition.

One of the most attractive technologies to reach the final goal of net zero emissions by 2050 lies in the use of green hydrogen that can be supplied to fuel cells for producing electricity and heat. Nowadays, airports are responsible for 13% of the European Union’s transport sector greenhouse gas emissions. In this paper, an innovative containerized modular trigeneration system, named “Hydro-Gen”, has been proposed to cover electric, thermal and cooling demands of a small-medium scale airport via fuel cells fuelled by green hydrogen. A dynamic simulation model of the “Hydro-Gen” system has been developed by means of the TRNSYS platform. The proposed system has been simulated under two operating scenarios with reference to a 1-year period while coupled with the selected airport demand profiles. The simulation results have been analyzed from energy and economic points of view and compared with a traditional energy generation scenario (where the central power grid only is used). The results underlined that the proposed system significantly reduces primary energy consumption under both scenarios up to a maximum 100.9%, while the economic performance are strongly dependent on the unit cost of hydrogen.

This research numerically investigates entropy generation in a laminar forced convective flow of Al$_2$O$_3$-water nanofluid within a 2D axisymmetric pipe under uniform heat flux. The study employed aluminum oxide nanoparticles with a consistent diameter of 30nm, dispersed in water at volumetric concentrations of 1% to 4%. For Reynolds numbers (Re) ranging from 200 to 900, the analysis focused on the interplay between the Nusselt number, thermal-hydraulic performance, and entropy generation as functions of both flow velocity and nanoparticle concentration. Results show that elevating the Re and volume fraction not only increases the Nusselt number but also reduces the total entropy generation by 22.25%. A corresponding rise in pressure drop was also observed with these increases. Consequently, the application of Al$_2$O$_3$-water nanofluid proves to be thermodynamically advantageous, enhancing heat transfer characteristics while simultaneously suppressing entropy generation.

Heating elements in a circular cross section are mostly utilized as the foundation of heat exchangers because of their simplicity and low production cost. Streamlined circular heaters may separate and create significant wakes, which can result in high-pressure drops. So, they have a much lower hydraulic area and thus need less pumping power. In the context of this research, the main objective is to experimentally investigate the hydrothermal parameters in a water tank heat exchanger heated by new heating fluid supply method. Several heat fluxes were studied during the experiment, and several parameters were considered, such as surface temperature, pumping power, heat flux, Reynolds number, and the averageNusselt. The correlations between those parameters have been developed and analyzed. The values of the Nusselt number change at any change in the Reynolds number, the power of tubular heaters, or the place of the heated cylinder inside the tank. A quasi-linear relationship between pumping power and pressure drop shows that for all heated cylinders in the Re range of 154.34 ≤ ReD ≤ 212.51, the range of mean pumping power was 0.54 × 10⁻⁴ ≤ Wp ≤ 4.9 × 10⁻⁴.

Accurate prediction of Standard Penetration Test (SPT) blow counts from Cone Penetration Test (CPT) data is critical for reliable geotechnical characterization, particularly when SPT data are scarce or difficult to obtain. This study presents a data-driven framework that employs an Artificial Neural Network (ANN) to estimate the corrected SPT blow number ($N_{60}$) using key CPT parameters. The database was compiled from two construction sites in Nasiriyah, Iraq, comprising cone tip resistance ($q_c$), sleeve friction ($fs$), and effective overburden pressure ($\sigma_{vo}^{\prime}$) as input variables. Multiple ANN architectures were trained and validated, and optimal performance was achieved using one hidden layer with eight neurons, yielding a coefficient of determination ($R^2$) of 0.9967, and two hidden layers with six and sixteen neurons, achieving $R^2$ = 0.9976. Relative importance analysis indicated that cone tip resistance contributed 44% to the model’s predictive strength, followed by sleeve friction and effective overburden pressure, each accounting for approximately 26%. Sensitivity analysis confirmed that $N_{60}$ increases with higher input parameters, consistent with soil behavior principles. The ANN model demonstrated high accuracy and generalization capability across both sandy and clayey soils. Design charts derived from the trained model enable practical estimation of $SPT-N$ from CPT results, providing geotechnical engineers with a rapid and reliable tool for site characterization and preliminary design.

Axial micro-hydro turbines' performance is sensitive to design and operational parameters. This study investigates the performance of micro-hydro turbines through advanced computational fluid dynamics simulations utilizing ANSYS FLUENT. A three-dimensional, steady-state RANS approach with the SST $k-\omega$ turbulence model was employed to simulate fluid flow interactions in turbines of 3, 4, and 5 blades, multiple rotational speeds of 100 and 1000 RPM, and flow rates. The research uniquely investigates a design that integrates internal flow control features by incorporating secondary guide blades and axial flow straighteners, which effectively reduce vortex formation and energy dissipation. Results show that increasing blade count significantly boosts mechanical power output and hydraulic head across flow conditions. Turbines operating at higher rotational speeds demonstrate markedly enhanced power generation, indicating the importance of mechanical design for durability under elevated stress. Comparative analyses between water and oil as working fluids reveal interesting fluid-dependent performance trends, with oil exhibiting superior energy transfer at higher RPMs. Validation against empirical data confirms the computational procedure's accuracy, with RMSE below 4%. The best performance was observed with the 5-blade turbine, reaching 95%, 92%, and 95% at rotational speeds of 100, 500, and 1000 RPM, respectively, at 60,700 Reynolds number. This configuration consistently outperformed others, demonstrating superior energy conversion. The addition of axial flow straighteners slightly improved efficiency by minimizing turbulence and vortex losses, confirming that structural enhancements combined with high rotational speeds are key to achieving maximum turbine performance.

This paper presents simulation and experimental validation of a Nonlinear Compensation-based Neuro-Fuzzy (NCNF) controller designed to balance the rotary inverted pendulum (RIP). Traditional linear controllers, such as Proportional-Integral-Derivative (PID) and state-feedback with pole placement, usually achieve satisfactory results in simulations on linearized models. However, their performance decreases in hardware implementation because of disturbances and unmodeled nonlinear effects such as Coulomb friction and mechanical backlash. To overcome these challenges, a feedforward compensation function was developed to cancel these undesired effects, which is combined with an Adaptive Neuro-Fuzzy Inference System (ANFIS) controller that updates PID gains to improve the rotary arm tracking for a square-wave reference and stabilize the pendulum at the upright position. The proposed NCNF controller is validated through hardware-in-the-loop (HIL) experiments and compared with a baseline state-feedback controller. Results show that the arm angle ($\theta$) overshoot decreased from 40.6% to 0.8% (lower step) and from 17.2% to 2.5% (upper), total steady-state $\theta$-error from 5.75° to 0.296°, and the fitness index dropped from 41.12 to 25.23. The nonlinear compensation reduced the gap between simulation and real-time performance, while the ANFIS further improved the defined control metrics. Overall, the NCNF controller achieves more stable and precise tracking than the state-feedback control.