Search

This computational study explores the potential of several soft-computing techniques for wind turbine (WT) output power (kW) estimation based on seven input variables of wind speed (m/s), wind direction (°), air temperature (°C), pitch angle (°), generator temperature (°C), rotating speed of the generator (rpm), and voltage of the network (V). In the present analysis, a nonlinear regression-based model (NRM), three decision tree-based methods (random forest (RF), random tree (RT), and reduced error pruning tree (REPT) models), and multilayer perceptron-based soft-computing approach (artificial neural network (ANN) model) were simultaneously implemented for the first time in the prediction of WT output power (WTOP). To identify the top-performing soft computing technique, the applied models' predictive success was compared using over 30 distinct statistical goodness-of-fit parameters. The performance assessment indices corroborated the superiority of the RF-based model over other data-intelligent models in predicting WTOP. It was seen from the results that the proposed RF-based model obtained the narrowest uncertainty bands and the lowest quantities of increased uncertainty values across all sets. Although the determination coefficient values of all competitive decision tree-based models were satisfactory, the lower percentile deviations and higher overall accuracy score of the RF-based model indicated its superior performance and higher accuracy over other competitive approaches. The generator's rotational speed was shown to be the most useful parameter for RF-based model prediction of WTOP, according to a sensitivity study. This study highlighted the significance and capability of the implemented soft-computing strategy for better management and reliable operation of wind farms in wind energy forecasting. [ABSTRACT FROM AUTHOR]

A rapid and efficient method for the simultaneous monitoring and recovery of Auramine O (AO) and Methylene Blue (MB) dyes from water samples is presented. This method, named ultrasound-assisted dispersive-magnetic nanocomposites-solid-phase microextraction (UA-DMN-μSPE), utilizes NiCoMn/Fe 3 O 4 @C composite sorbents. Response surface methodology (RSM) combined with artificial neural networks (ANN) and generalized regression artificial neural network (GRNN) under central composite design (CCD) was employed to optimize various parameters for efficient extraction, followed by further refinement using desirability function analysis (DFA) and genetic algorithms (GA). Under optimized conditions, the method achieved exceptional recovery rates (99.5 ± 1.2% for AO and 99.8 ± 1.1% for MB) with acetone as the eluent. Additionally, a high preconcentration factor of 45.50 and 47.30 for AO and MB, respectively, was obtained. Low detection limits of 0.45 ng mL⁻1 (AO) and 1.80 ng mL⁻1 (MB) were achieved with wide linear response ranges (5–1000 and 5–2000 ng mL⁻1 for AO and MB, respectively). The method exhibited good stability with RSDs below 3% for five recycling runs, and minimal interference from various ions was observed. This UA-DMN-μSPE-UV/Vis method offers significant advantages in terms of efficiency, preconcentration, and detection limits, making it a valuable tool for the analysis of AO and MB in water samples. [Display omitted] • UA-DMN-μSPE method optimized using RSM, ANN, and GRNN ensures high-performance MB and AO extraction. • Optimized UA-DMN-μSPE-UV/Vis method, developed with DFA and GA, enhances recovery, preconcentration, and selectivity for MB and AO. • Concurrent monitoring & recovery offers sustainable solution for dye removal. • Potential to address environmental & health concerns from non-degradable dyes. [ABSTRACT FROM AUTHOR]

In the present paper, the removal of ethidium bromide (EtBr) from aqueous solutions in a batch system using natural (NP) and aluminium-coated pumice (ACP) as alternative low-cost adsorbents was investigated. The maximum adsorption capacity, q m (mg/g) was 58.82 and 76.92 mg/g for NP and ACP, respectively, operating at an initial pH of 8, an adsorbent dose of 8 g/L, a contact time of 210 min and an initial EtBr concentration of 30 mg/L. The equilibrium data of both adsorbents fitted the Freundlich isotherm model, indicating the heterogeneity of the adsorbent surface. In addition, the adsorption rate of both adsorbents was well described by the pseudo-second-order kinetics model. This indicated chemisorption was the rate-controlling step of the adsorption process which occurred by ion exchange. Within the performed study, a three-layer artificial neural network (ANN) model was also developed to predict the efficiency of EtBr removal. Computational results clearly demonstrated that the ANN model was able to predict the combined effect of initial pH, adsorbent dose, contact time and initial EtBr concentration on the adsorption efficiency with a very high determination coefficient ( R 2 = 0.998) and a low relative error (RE = 0.037). [ABSTRACT FROM AUTHOR]

The efficiency of new and nano-scale adsorbents including multi-walled carbon nanotubes (MWCNTs) and γ-alumina in the removal of cobalt(II) from aqueous solutions was experimentally evaluated in a batch-system reactor. To the best of our knowledge, no previous study has specifically attempted to introduce a hybrid strategy based on artificial neural network and genetic algorithm techniques for modelling and optimizing adsorptive removal of cobalt(II) from aqueous solutions via the proposed nanoparticles. The analyses of SEM, TEM, and FTIR were used to characterize both adsorbents. The response surface methodology (RSM) approach suggested a second-order polynomial model with a p -value < 0.0001 and R 2 of 0.9980 for MWCNTs adsorbent and a p -value < 0.0001 and R 2 of 0.9992 for γ-alumina adsorbent. The artificial neural network (ANN) approach suggested a three-layered feed-forward backpropagation model with R 2 of 0.9794 for MWCNTs adsorbent and R 2 of 0.9823 for γ-alumina adsorbent. The results linked to optimization by RSM showed that the maximum cobalt(II) removal efficiency of about 90% was achieved in the case of the MWCNTs adsorbent under the conditions of pH = 10, contact time = 38.6 min, MWCNTs dosage = 1.57 mg/L, and initial cobalt(II) concentration = 56.57 mg/L. About 93% of cobalt(II) removal could be obtained in the case of γ-alumina adsorbent under the conditions of pH = 10, contact time = 35.5 min, γ-alumina dosage = 1.63 g/L, and initial cobalt(II) concentration = 52.15 mg/L. The optimization values using the genetic algorithm (GA) technique were almost the same as those obtained from the RSM method. The kinetic model of Ho and McKay's pseudo-second order (PSO) and the isotherm model of Dubinin–Radushkevich were found to be the best-fitted to the experimental for both MWCNTs and γ-alumina. In addition, the maximum monolayer adsorption capacity of MWCNTs and γ-alumina adsorbents for the adsorption of cobalt(II) was 78.94 mg/g and 75.78 mg/g, respectively. Also, a thermodynamic study exhibited a favorable and spontaneous adsorption process for both materials. The present study clearly concluded that the proposed adsorbents could be effectively used for the removal of cobalt(II) from aqueous solutions at lower adsorbent dose and shorter contact times than various adsorbents reported in literature. Unlabelled Image • Adsorptive removal of cobalt (II) was explored via MWCNTs and nano-alumina. • ANN and genetic algorithm were proposed for this first time in cobalt (II) adsorption. • Isotherms and kinetic data followed Dubinin–Radushkevich and PSO models, respectively. • A cobalt (II) removal efficiency above 90% was possible at pH = 10 and 1 g adsorbent/L. • Favorable and spontaneous removal was corroborated by thermodynamic study. [ABSTRACT FROM AUTHOR]