Sign in Subscribe
Energy management

Machine Learning-Based Framework Improves Sustainability in Greenhouse Systems

Strategic Insights

— 1 min read

Researchers at Sichuan University in China have developed a machine learning-based framework to precisely predict power consumption and freshwater production in greenhouse-integrated systems. The framework has shown significant improvements in resource efficiency and sustainable agriculture, particularly in areas with limited water resources or high energy consumption. By combining machine learning and multi-objective optimization, the researchers have created a tool that can design greenhouse systems with optimal freshwater production and energy consumption. The findings of this study have been published in the journal Sustainable Computing-informatics & Systems.

Key Takeaways:

  • The traditional modeling techniques frequently fail to address the complex, multi-variable optimization problem in greenhouse systems, which is essential for improving the sustainability of controlled agricultural settings.
  • The machine learning-based framework developed in this study suggests a strong framework for precisely predicting power consumption and freshwater production in greenhouse-integrated systems.
  • Five-fold cross-validation, hybrid Grey Wolf Optimizer (GWO) tuning, SHAP sensitivity analysis, and Taylor diagrams were used to assess various machine learning models, such as XGBoost, CatBoost, SVR, MLP, KNN, and ElasticNet.
  • The XGBoost-GWO model outperformed the others, obtaining the highest R2 values (up to 0.9991) and the lowest RMSE (0.4933 for freshwater, 0.0311 for power).
  • Greenhouse width was found to be the most significant design parameter by feature importance and sensitivity analyses.
  • An ideal configuration that produced 99.80 m3 of freshwater per day with a mere 2.75 kWh/m3 energy consumption was found using a multi-objective optimization approach.
  • The combined modeling and optimization method promotes resource efficiency and sustainable agriculture by providing a useful tool for designing greenhouse systems.

Statistics:

  • 99.80 m3 of freshwater per day
  • 2.75 kWh/m3 energy consumption
  • Up to 0.9991 R2 value
  • 0.4933 RMSE for freshwater
  • 0.0311 RMSE for power
  • 5-fold cross-validation
  • Hybrid Grey Wolf Optimizer (GWO) tuning
  • SHAP sensitivity analysis
  • Taylor diagrams

Sources:

  • "Machine Learning-based Model for Predicting Freshwater Production and Power Consumption In Solar-assisted Desalination Systems. Sustainable Computing-informatics & Systems, 2025;47."
  • Sustainable Computing-informatics & Systems, Elsevier
  • Radarweg 29, 1043 Nx Amsterdam, Netherlands
  • "Investigators at Sichuan University Describe Findings in Machine Learning (Machine Learning-based Model for Predicting Freshwater Production and Power Consumption In Solar-assisted Desalination Systems). Journal of Engineering. October 20, 2025; p 1247."

Read more