Performance assessment of waste heat/solar driven membrane-based simultaneous desalination and liquid desiccant regeneration system using a thermal model and KNN machine learning tool

In this work, the waste heat/solar heat-driven membrane-based liquid desiccant regenerator performance, as well as desalinated water extraction rate, are predicted and analyzed by developing a thermal model and KNN-ML tool. In the membrane-based liquid desiccant regenerator, water is used as a working fluid instead of scavenging air for desalinated water extraction purpose. The proposed thermal model and KNN-ML tool are validated with the literature data and found in good agreement. Optimal inlet conditions were determined for the given operating range using the thermal model and KNN-ML tool. With vapor flux and energy exchange as the performance indicators, the water extraction rate and thermal performance of the membrane-based liquid desiccant regenerator are predicted for the optimal inlet condition using the KNN-ML tool. Also, the heat and mass transfer characteristics such as Lewis number and NTUm across the membrane in the liquid desiccant regenerator are assessed using the developed thermal model. Further, for the optimal inlet conditions, utilizing waste heat from thermal power plants (Method-I) and solar energy from solar heater (Method-II), the thermal performance and water extraction rate across the membrane in the liquid desiccant regenerator are assessed based on the developed thermal model.