Experimental Evaluation and Development of Artificial Neural Network Model for the Solar Stills Augmented with the Permanent Magnet and Sandbag

Abstract

The availability of potable water is reducing day by day due to rapid growth in the human population and un-planned industrialization around the globe. Although human beings cannot think of survival in the absence of water, the global leadership can still not implement their pacts in reality. Solar still is one of the prominent ways of getting potable water from contaminated water. This manuscript reports the experimental evaluation and developed ANN model for the single basin solar stills having augmentations with the sand-filled cotton bags and ferrite ring permanent magnets. Root mean square error (RMSE), efficiency coefficient (E), the overall index of model performance (OI), and coefficient of residual mass (CRM) values are in good agreement with the proposed developed model of ANN. The proposed ANN model can be utilized to predict distillate yield with a variation of 5% for the reported modified stills. Overall correlation coefficient of CSS, MSS-1&2 are 0.98171, 0.9867, and 0.99542, respectively.