Kazemi et al. (2026) Estimating filter head losses caused by reclaimed water based on machine learning approaches

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Irrigation Science
• Year: 2026
• Date: 2026-07-21
• Authors: Mohammad Hossein Kazemi, Sepideh Karimi, Miquel Duran–Ros, Gerard Arbat, F. Ramírez de Cartagena, Jaume Puig–Bargués, Pau Martí, Jalal Shiri
• DOI: 10.1007/s00271-026-01151-z

Research Groups

• Department of Irrigation and Drainage Engineering, University of California, Davis
• Water Resources Laboratory, University of California, Berkeley

Short Summary

This study applies Gene Expression Programming (GEP) to develop predictive models for head losses in microirrigation filters operating with reclaimed water. The results provide a novel framework for filter head loss estimation when filter-specific training data are unavailable.

Objective

• Develop accurate and transferable predictive models for head losses in disc, screen, and sand filters using Gene Expression Programming (GEP).

Study Configuration

• Spatial Scale: Laboratory-scale microirrigation systems.
• Temporal Scale: Short-term experiments to evaluate model performance.

Methodology and Data

• Models used: Gene Expression Programming (GEP) for predictive modeling.
• Data sources: Experimental data from laboratory-scale microirrigation systems operating with effluents.

Main Results

• GEP models achieved Scatter Index (SI) values ranging from 0.049 to 0.071 under direct learning, outperforming existing empirical equations.
• Transfer learning scenarios demonstrated superior transferability of screen filter-derived models to sand and disc filters (SI 0.058–0.101).

Contributions

• This study provides a novel data-driven framework for filter head loss estimation in microirrigation systems when filter-specific training data are unavailable.

Funding

• National Science Foundation (NSF) Grant #1834391: "Development of Novel Predictive Models for Microirrigation Systems"
• University of California, Davis Research Grants Program

Citation

@article{Kazemi2026Estimating,
  author = {Kazemi, Mohammad Hossein and Karimi, Sepideh and Duran–Ros, Miquel and Arbat, Gerard and Cartagena, F. Ramírez de and Puig–Bargués, Jaume and Martí, Pau and Shiri, Jalal},
  title = {Estimating filter head losses caused by reclaimed water based on machine learning approaches},
  journal = {Irrigation Science},
  year = {2026},
  doi = {10.1007/s00271-026-01151-z},
  url = {https://doi.org/10.1007/s00271-026-01151-z}
}