Modeling of energy efficiency for a solar still fitted with thermoelectric modules by ANFIS and PSO-enhanced neural network: A nanofluid application

An accurate model is developed for predicting the energy efficiency of a single-slope solar still equipped with thermoelectric modules. Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN) enhanced by Particle Swarm Optimization (PSO) are employed. Cu2O nanopartides are utilized in the solar still basin, and the energy efficiency is modeled as a function of the time, glass temperature, fan power, solar radiation, ambient temperature, water temperature, basin temperature as well as the nanopartide volume fraction. The experimental data arc utilized for training the artificial intelligence methods. The ANN with three hidden neurons and the ANTIS with nine clusters present the best predictions. Applying the PSO profoundly enhances the prediction performance. The comparison between the performances of PSO-based ensemble models reveals superiority of the PSO-ANFIS compared with the PSO-ANN. The R-2 values for the PSO-ANFIS model are 0.9884 and 0.9906 for the training and test sets, respectively. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

An accurate model for predicting the energy efficiency of a single-slope solar still equipped with thermoelectric modules is developed using ANFIS and PSO-enhanced ANN. Cu2O nanoparticles are utilized in the basin, and experimental data is used for training the models. PSO-ANFIS model shows superior prediction performance compared to PSO-ANN.