Soft Computing Techniques for Classification of Voiced/Unvoiced Phonemes

A method that uses fuzzy logic to classify two simple speech features for the automatic classification of voiced and unvoiced phonemes is proposed. In addition, two variants, in which soft computing techniques are used to enhance the performance of fuzzy logic by tuning the parameters of the membership functions, are also presented. The three methods, manually constructed fuzzy logic (VUFL), fuzzy logic optimized with genetic algorithm (VUFL-GA), and fuzzy logic with optimized particle swarm optimization (VUFL-PSO), are implemented and then evaluated using the TIMIT speech corpus. Performance is evaluated using the TIMIT database in both clean and noisy environments. Four different noise types from the AURORA database—babble, white, restaurant, and car noise—at six different signal-to-noise ratios (SNRs) are used. In all cases, the optimized fuzzy logic methods (VUFLGA and VUFL-PSO) outperformed manual fuzzy logic (VUFL). The proposed method and variants are suitable for applications featuring the presence of highly noisy environments. In addition, classification accuracy by gender is also studied.     

Pseudo bond graph model of thermal transfers sustained by ice quantity of a domestic refrigerator for energy saving application

This paper presents a study on the use of ice to improve the energy efficiency of a domestic refrigerator by applying a pseudo bond graph model that describes the thermal transfers sustained by a quantity of ice introduced inside the cavity of refrigeration. The use of ice resulted in a global energy saving of 4.68%. The effect of ice was found to be more significant during the transitional regime. It reduced the response time to reach the stable average temperature from 15 h to only 3.5 h compared to when not using ice. This achievement did not cost additional electrical power, but rather allowed a saving of electrical energy of 76.73%. However, during the steady state, a reduction in the energy efficiency was noted. An improvement in the cooling by keeping the temperature inside the refrigerator more homogeneous is also proved. The model has two inputs: the outside temperature, and the modulated temperature of the evaporator. This latter determines the functioning of the compressor cycle. The model describes the thermal transfers by natural convection inside the refrigerator. Two experiments were carried out to make a performance comparison and to prove the influence of ice in cooling and energy saving. We used real measurements to modulate the evaporator temperature source in the pseudo bond graph model. The simulation results show the effectiveness of the proposed approach. © 2015 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Smart and Multifunctional Fiber-Reinforced Composites of 2D Heterostructure-Based Textiles

Smart and multifunctional fiber reinforced polymer (FRP) composites with energy storage, sensing, and heating capabilities have gained significant interest for automotive, civil, and aerospace applications. However, achieving smart and multifunctional capabilities in an FRP composite while maintaining desired mechanical properties remains challenging. Here, a novel approach for layer-by-layer (LBL) deposition of 2D material (graphene and molybdenum disulfide, MoS₂)-based heterostructure onto glass fiber fabric using a highly scalable manufacturing technique at a remarkable speed of ≈150 m min⁻¹ is reported. This process enables the creation of smart textiles with integrated energy storage, sensing, and heating functionalities. This methodology combines gel-based electrolyte with a vacuum resin infusion technique, resulting in an efficient and stable smart FRP composite with an areal capacitance of up to ≈182 µF cm⁻² at 10 mV s⁻¹. The composite exhibits exceptional cyclic stability, maintaining ≈90% capacitance after 1000 cycles. Moreover, the smart composite demonstrates joule heating, reaching from ≈24 to ≈27 °C within 120 s at 25 V. Additionally, the smart composite displays strain sensitivity by altering electrical resistance with longitudinal strain, enabling structural health monitoring. These findings highlight the potential of smart composites for multifunctional applications and provide an important step toward realizing their actual real-world applications.     

Double diffusive Buoyancy-driven flow in a fluid-saturated elliptical annulus with a neural network-based prediction of heat and mass transfer

This paper presents a numerical study of buoyancy-driven double-diffusive convection within an elliptical annulus enclosure filled with a saturated porous medium. An in-house built FORTRAN code has been developed, and computations are carried out in a range of values of Darcy–Rayleigh number Ra_m (10 ≤ Ra_m ≤ 500), Lewis number Le (0.1 ≤ Le ≤ 10), and the ratio of buoyancy forces N (−5 ≤ N ≤ 5). In addition, three methods are used, namely the multi-variable polynomial regression, the group method of data handling (GMDH), and the artificial neural network (ANN) for the predictions of heat and mass transfer rates. First, results are successfully validated with existing numerical and experimental data. Then, the results indicated that temperature and concentration distributions are sensitive to the Lewis number and thermal and mass plumes are developing in proportion to the Lewis number. Two particular values of Lewis number Le = 2.735 and Le = 2.75 captured the flow's transition toward an asymmetric structure with a bifurcation of convective cells. The average Nusselt number tends to have an almost asymptotic value for Le » 5. For the case of aiding buoyancies N > 1, the average Nusselt Number $\overline{\mathit{Nu}}$ decreased by 33% when the Lewis number increased to its maximum value. Then, it increased by 10% when the Lewis number increased to Le = 1 for the case of opposing buoyancies N < 1 and then decreased by 33% when the Lewis number increased to its maximum value., contrary to the behavior of the average Sherwood number $\overline{\mathit{Sh}}$ that increased by 700% for both cases N > 1 and N < 1. New correlations of $\overline{\mathit{Nu}}$, and $\overline{\mathit{Sh}}$ as a function of Ra_m, Le, and N are derived and compared with GMDH and ANN methods, and the ANN method showed higher performance for the prediction of $\overline{\mathit{Nu}}$ and $\overline{\mathit{Sh}}$ with R² exceeding 0.99.     

Synthesis,Characterization and Adsorption of Bisphenol A Using Novel Hybrid Materiel Produced from PANI Matrix Reinforced by Kieselguhr

Journal of Inorganic and Organometallic Polymers and Materials - This study investigated the preparation of new adsorbent based PANI and Kieselguhr (KG). The produced materials were characterized...     

CADS-ML/DL: efficient cloud-based multi-attack detection system

International Journal of Information Security - With the increasing adoption of cloud computing, securing cloud-based systems and applications has become a critical concern for almost every...