Design of solar power-based hybrid desalination predictive method using optimized neural network

A hybrid solar-powered desalination system is one of the potential energy-intensive solutions that can be utilized to meet the increasing demand for freshwater. Modern desalination methods rely heavily on using artificial intelligence (AI) methodologies. This paper proposes a novel and optimal approach using neural networks to predict the effectiveness of a hybrid solar-powered desalination system. The experimental design furnishes the requisite data for conducting the analysis. The appropriate features are then retrieved and chosen using the VGG16 approach. Therefore, it is suggested to use the reptile search optimization with radial basis long short-term memory (RSO-RBLSTM) approach to forecast how well the hybrid solar-powered desalination process would work. Using the RSO method, the hyperparameters of the suggested algorithm are tuned. This study's implementation makes use of the MATLAB programming language. The simulation results are verified using many traditional techniques using various metrics, including RMSE, MAE, MSE, R2, and AARD. The proposed approach has attained 0.05 RMSE, 0.023 MSE, 0.06 MAE, and 4.5 % AARD for the training set. The proposed method has achieved 0.056 RMSE, 0.01 MSE, 0.025 MAE, and 4.2 % AARD for the testing set. The metrics value displays the hybrid desalination system's effective performance when powered by the solar system.

Automated summary

This paper proposes a novel approach using neural networks to predict the effectiveness of a hybrid solar-powered desalination system and verifies it using simulation results. The RSO-RBLSTM method is utilized to forecast the performance of the hybrid desalination process, and the proposed approach shows effective performance when powered by the solar system according to various metrics.