Project Description

The freshwater crisis presents an intimidating global challenge, particularly within the developing regions of the Middle East and coastal South Asia. Desalination stands out as a promising technological solution to these water needs by removing salts and impurities from seawater. However, the diverse matrix of desalination technologies requires complex, multidirectional efforts to balance freshwater production with rising environmental and energy concerns. Conventional thermal desalination remains energy-intensive, while membrane-based technologies often struggle with low water recovery ratios and membrane fouling.

Humidification-Dehumidification (HDH) technology addresses these gaps by mimicking the natural water cycle. In this process, air is humidified by flowing over warm seawater and subsequently cooled to achieve the condensation of high-purity freshwater. In hot and humid regions, such as the UAE and coastal India, HDH has immense potential due to the natural availability of humid air and abundant solar radiation. By integrating solar energy, desalination can be achieved with reduced reliance on fossil fuels while simultaneously providing localized thermal comfort through air dehumidification.

This Ph.D. research focuses on both closed and open-cycle HDH systems optimized for simultaneous water production and indoor thermal comfort. The process of space conditioning is achieved by utilizing the dehumidifier as a cooling and dehumidification coil. As the hot, saturated air from the humidifier enters the dehumidifier, it encounters a cooling medium. This interaction triggers two critical processes: Latent Heat Removal (Water vapor condenses into freshwater, significantly reducing the humidity of the air stream) and Sensible Cooling (The air temperature is lowered to a desired dry-bulb setpoint, making it suitable for indoor climate control).

To transcend the limitations of traditional thermodynamic modeling, this research integrates Machine Learning (ML) to drive system optimization. ML interventions will be utilized in three primary capacities: Multi-Objective Design Optimization (Using algorithms such as Genetic Algorithms (GA) or Particle Swarm Optimization (PSO), the system can identify the optimal geometric configurations (e.g., packing density, heat exchanger surface area) that maximize both the Gained Output Ratio (GOR) and the Cooling Capacity simultaneously), Performance Prediction (Artificial Neural Networks (ANN) will be trained on experimental and simulated data to predict water productivity and thermal comfort levels under fluctuating ambient conditions, such as varying solar intensity and inlet air humidity) and Real-Time Operational Control (ML-based controllers can adjust mass flow rate ratios in real-time to maintain stable indoor temperatures while ensuring the desalination process operates at peak exergy efficiency).

The research will involve rigorous modeling and experimentation to enhance water productivity and cooling capacity by optimizing operating conditions and selecting advanced material characteristics. Furthermore, the project includes the sampling of seawater at various locations to determine pre-treatment requirements, allowing for an accurate estimation of the Levelized Cost of Water (LCOW).

Ultimately, this project aims to provide a portable solution for decentralized freshwater generation and climate control in enclosed environments, alongside a robust framework for scaling this technology for community and industrial applications.