SUMMARY

Water desalination is a reliable process for supplying freshwater water, but geography, cost (operating/capitol), and environmental impacts pose significant challenges for widespread adoption. Today, nearly all (∼99%) desalination plants rely on fossil fuels as the primary energy source to produce heat or electricity to drive the desalination process. If this trend continues, carbon emissions from fossil fuel-powered desalination plants could increase to 400 million tons of CO2 per year by 2050. Additionally, the projected waste brine produced at current desalination plants by 2050 may approach 240 km3 per year. Solar-thermal desalination technologies could provide a sustainable path toward achieving high volume (million gallons per day) renewable driven desalted water while achieving minimal or zero liquid discharge (MLD and ZLD). This is of growing interest in an effort to minimize waste (carbon and brine). Yet, efficient integration between solar capture and desalination remains as critical challenge. Furthermore, the high energy intensity required to reach saturation in MLD/ZLD processes is a critical obstacle. The aim of this PhD proposal and dissertation is to propose a framework that integrates thermodynamics, geographical information systems, and machine learning for developing and predicting sustainable strategies to integrate solar energy with desalination. Using computational models and data analysis, this research program aims to evaluate the current state of desalination worldwide and the potential of solar thermal hybrid desalination systems. The use of machine learning benefits the processing of high amounts of data and system optimization, decreasing the computational time and resource consumption.