A smart simulation-optimization framework for solar-powered desalination systems

Abstract

Energy crises caused some secondary challenges in the field of water and food as well its inherent catastrophes. In the nexuses of water and energy, supplying the power for desalination and water treatment process is assumed as an important subject in scientific communities. While, in sunny arid and semi-arid climates (Zahedan, Iran), the application of solar irradiances can convert threats into opportunities. In the present research, in the first step, all operational features in photovoltaic facilities are optimized by Central Composition Design (CCD) as a Response Surface Methodology (RSM) method. In the second step, harvested renewable energy values, carbon emission prevention, initial cost, and maintenance for each year are predicted as per operational factors of the photovoltaic system, which are utilized in the desalination process and with the application of Gaussian Processes, Multilayer Perceptron, Support Vector Machine for Regression (SMOreg) and Polynomial Trees (M5P) algorithms. Finally, in the managerial insight section of the research, an integrated Reverse Osmosis-Photovoltaic system is presented as a green facility of water providing in arid and semi-arid climates. Based on RSM system the main effective factors for energy production, prevention carbon emission (25 year), initial and maintenance costs are Pitch degree (P) (P-value < 0.0001), Pitch degree (P-value < 0.0001), Cross Distance (CD) (P-value < 0.0001), and Panel Capacity (PC) (P-value = 0.0002), respectively. Finally, an SDGS assessment is performed in this investigation, and some managerial insights are proposed. In the different steps of this study, both carbon emission values and cost functions are optimized. While, to the prediction of available solar energy, Gaussian process with more than 0.98 has the most efficiency.