RESUMO

This study investigates the application of the Gaussian Radial Basis Function Neural Network (GRNN), Gaussian Process Regression (GPR), and Multilayer Perceptron Optimized by Particle Swarm Optimization (MLP-PSO) models in analyzing the relationship between rainfall and runoff and in predicting runoff discharge. These models utilize autoregressive input vectors based on daily-observed TRMM rainfall and TMR inflow data. The performance evaluation of each model is conducted using statistical measures to compare their effectiveness in capturing the complex relationships between input and output variables. The results consistently demonstrate that the MLP-PSO model outperforms the GRNN and GPR models, achieving the lowest root mean square error (RMSE) across multiple input combinations. Furthermore, the study explores the application of the Empirical Mode Decomposition-Hilbert-Huang Transform (EMD-HHT) in conjunction with the GPR and MLP-PSO models. This combination yields promising results in streamflow prediction, with the MLP-PSO-EMD model exhibiting superior accuracy compared to the GPR-EMD model. The incorporation of different components into the MLP-PSO-EMD model significantly improves its accuracy. Among the presented scenarios, Model M4, which incorporates the simplest components, emerges as the most favorable choice due to its lowest RMSE values. Comparisons with other models reported in the literature further underscore the effectiveness of the MLP-PSO-EMD model in streamflow prediction. This study offers valuable insights into the selection and performance of different models for rainfall-runoff analysis and prediction.

RESUMO

A hybrid neural model (HNM) and particle swarm optimization (PSO) was used to optimize ethanol production by a flocculating yeast, grown on cashew apple juice. HNM was obtained by combining artificial neural network (ANN), which predicted reaction specific rates, to mass balance equations for substrate (S), product and biomass (X) concentration, being an alternative method for predicting the behavior of complex systems. ANNs training was conducted using an experimental set of data of X and S, temperature and stirring speed. The HNM was statistically validated against a new dataset, being capable of representing the system behavior. The model was optimized based on a multiobjective function relating efficiency and productivity by applying the PSO. Optimal estimated conditions were: S0 = 127 g L-1, X0 = 5.8 g L-1, 35 °C and 111 rpm. In this condition, an efficiency of 91.5% with a productivity of 8.0 g L-1 h-1 was obtained at approximately 7 h of fermentation.

Assuntos

RESUMO

ABSTRACT The primary challenge in organizing sensor networks is energy efficacy. This requisite for energy efficacy is because sensor nodes capacities are limited and replacing them is not viable. This restriction further decreases network lifetime. Node lifetime varies depending on the requisites expected of its battery. Hence, primary element in constructing sensor networks is resilience to deal with decreasing lifetime of all sensor nodes. Various network infrastructures as well as their routing protocols for reduction of power utilization as well as to prolong network lifetime are studied. After analysis, it is observed that network constructions that depend on clustering are the most effective methods in terms of power utilization. Clustering divides networks into inter-related clusters such that every cluster has several sensor nodes with a Cluster Head (CH) at its head. Sensor gathered information is transmitted to data processing centers through CH hierarchy in clustered environments. The current study utilizes Multi-Objective Particle Swarm Optimization (MOPSO)-Differential Evolution (DE) (MOPSO-DE) technique for optimizing clustering.