Soft robotics has emerged as a promising and rapidly evolving field, replacing traditional rigid systems with highly compliant elastomer actuators that offer adaptable interactions with their surroundings. Driven by cost-effective fabrication techniques like molding, 3D-printing, and laser cutting, these systems frequently utilize pneumatic actuators to mimic biological muscle behavior. However, optimizing these designs remains difficult due to the highly nonlinear mechanical behavior of rubber-like materials and the complex fluid-structure interactions that occur during pneumatic actuation. To address these challenges, the primary aim of this research is to analyze the mechanical behavior of laser-cut, pneumatically actuated multi-material soft robotic structures. To achieve this, finite element (FE) simulations, validated against experimental measurements, were used to evaluate how different geometries and material compositions respond to actuation. By systematically examining these mechanical responses, this research aims to contribute to a broader understanding of how soft robots can be deployed more effectively and widely in the future. The main results showed that the triangular actuator shape is optimal for low-pressure, high-precision applications, whereas the rectangular and elliptical actuators can generate high grasping forces. The results also highlight the importance of the internal design of the actuator, showing that increased chamber density enhances bending in certain shapes while causing localized distortions in others.
Ranking manufacturing processes has significant technical and economic importance in production operations. However, this task is intricate due to the inherent diversity of technical and economic parameters in different manufacturing processes. This complexity makes the ranking of manufacturing processes a Multi-Criteria Decision-Making (MCDM) problem. This study evaluated the application of the recently proposed method, named Deviation-Based Pairwise Assessment Ratio Technique (DEPART), to rank production processes across three distinct scenarios, including ranking nine alternatives in metal grinding using slotted grinding wheels, ranking nine metal turning processes, and ranking six extraction processes in the chemical field. In each example, the process ranking results obtained via the DEPART method were compared to those generated by six other MCDM methods, including Simple Additive Weighting (SAW), Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), Multiobjective Optimization On the basis of Ratio Analysis (MOORA), COmplex PRroportional Assessment (COPRAS), Root Assessment Method (RAM), and probability method. The findings indicated that the DEPART method is suitable for ranking production alternatives. However, upon reviewing all utilized MCDM techniques, the probability method emerged as the most appropriate method for ranking production alternatives within the analyzed problems.
This study examines the addition of silica sand nanoparticles at 12 wt.%, 22 wt.%, 32 wt.%, 42 wt.%, 52 wt.%, and 62 wt.% into epoxy resin matrices (DGEBA, Medapoxy HR EA, and Medapoxy HR EB) to enhance their physical, mechanical, and microstructural properties. In research, Silica was successfully incorporated up to 42 wt.%, exceeding the typical 25 wt.% inorganic threshold. This incorporation significantly boosts the compressive, flexural, and tensile strengths of the nanocomposites to 113.685 MPa, 49.723 MPa, and 33.452 MPa, respectively, showing increases of 57.49%, 40.33%, and 37.31% over pure epoxy. The optimal silica content for maximizing mechanical properties lies between 32 wt.% and 42 wt.%. Scanning electron microscopy (SEM) reveals that pure epoxy has a smooth and brittle fracture surface, whereas epoxy/SiO₂ composites exhibit cohesive deformations and fractures due to the silica. However, silica content above 42 wt.% leads to nanoparticle agglomeration, which reduces performance. The study concludes that silica nanoparticles improve the mechanical properties of epoxy resin, making the nanocomposites more economical, durable, and suitable for various industrial and domestic applications. These findings support the development of high-performance materials from abundant natural byproducts.
Spent fluid catalytic cracking (FCC) catalysts represent an environmental challenge. Due to their high Al₂O₃ and SiO₂ content, they have significant potential for reuse as raw materials and additives in ceramics. In this work, the sintering behaviour, phase evolution and densification of kaolin- based ceramics were studied. Compositions containing 30, 50, and 70 mass% spent FCC catalyst (designated as C30, C50, and C70) were developed. The pressed blends were fired at temperatures between 1250 and 1500°C. X‑ray diffraction results show that the C50 samples present the highest mullite content, while cristobalite dominates in the C70 composition. C30 composition achieved the best densification. At 1500°C, they reach an apparent density of 2.38 g/cm3, an apparent porosity of up to 5.4% and water absorption of 2.3%. The decrease in surface area of the C30 sample—from 58.9 m²/g in the raw state to 1.3 m²/g after firing at 1250°C—confirms effective sintering. These results show that the spent FCC catalyst is an effective additive in ceramic production.
This study focuses on short-term (10-minute) forecasting of Global Horizontal Irradiance (GHI) using artificial neural networks (ANNs) enhanced by three metaheuristic optimization algorithms: the FireFly Algorithm (FFA), Particle Swarm Optimization (PSO), and the White Whale Optimization Algorithm (WWOA). The models were trained using meteorological and astronomical data collected from a monitoring station in Khenchela, Algeria. To identify the solar radiation component most strongly correlated with GHI, one additional radiometric input, selected from the available components, was introduced in separate experiments. Model performance was assessed using standard statistical metrics: relative Root Mean Squared Error (rRMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). Results indicate that meteorological and astronomical variables alone are insufficient for highly accurate GHI prediction. However, when Global Tilted Irradiance (GTI) was incorporated as an additional input, all three hybrid models exhibited significantly improved accuracy. The ANN-FFA model achieved the best performance, with rRMSE=3.73%, MAPE=6.07%, and R²=0.9962. These findings demonstrate that GTI is the solar radiation component most closely related to GHI under the studied conditions. Furthermore, the study confirms the effectiveness of FFA, PSO, and WWOA in optimizing ANN hyperparameters for solar irradiance forecasting, with FFA yielding the most robust results.
