Work

Novel hybrid approach of random forest and stacking ensemble to improve fresh water yield prediction in mobile wick solar still

Year: 2025

Type: article

Abstract: Freshwater scarcity remains a global concern, and solar stills offer a sustainable solution, but they have low yields. This work employs machine learning to predict and optimize solar still productivity from environmental and operational data. Random Forest, AdaBoost, XGBoost, and Linear Regression were trained (80:20 train:test split) with GridSearchCV for hyperparameter tuning. Random Forest achieved the highest accuracy (train R 2 = 0.975, test R 2 = 0.830) and surpassed both XGBoost (test R 2 = 0.784) and the stacking ensemble (test R 2 = 0.783). Feature importance analysis identified solar still interior temperature and global irradiation as the most influential parameters. These findings highlight the potential of machine learning to enhance solar still performance. Future work will focus on integrating real-time environmental data and expanding the model deployment for adaptive freshwater production techniques.

Source: Desalination and Water Treatment

DOI: 10.1016/j.dwt.2025.101490

Authors: Sachinkumar Makwana, Dineshkumar Vaghela, Choon Kit Chan, Nithesh Naik

Institutions: Gujarat Technological University, Government Engineering College, Bhavnagar, INTI International University, Manipal Academy of Higher Education

Language: English

Topic: Solar-Powered Water Purification Methods

Subfield: Renewable Energy, Sustainability and the Environment

Field: Energy

Domain: Physical Sciences

SDGs: Affordable and clean energy, Clean water and sanitation, Zero hunger, Industry, innovation and infrastructure, Climate action

Open Access status: gold

Estimated APC paid: $1,000