Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 39 min ago
PreprintCC BYChemRxiv

Data-driven models for performance prediction in a microalgae-based wastewater treatment process

  • Ga-Yeong Kim1,2
  • JooMan Kim3,2
  • Jeremy S. Guest1,2,3
  1. 1Grainger College of Engineering, Department of Civil and Environmental Engineering
  2. 2University of Illinois Urbana–Champaign
  3. 3Institute for Sustainability, Energy, and Environment
Open at sourcePDFDOICite

Abstract

Microalgae-based wastewater treatment has the potential to advance tertiary phosphorus recovery while producing biomass for resource valorization. The development of reliable near-term prediction models can help reduce system upsets, support proactive process control, and optimize resource utilization. In this study, Light Gradient-Boosting Machine (LGBM), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) models were comparatively evaluated for predicting effluent ammonium, nitrate, and orthophosphate concentrations and biomass dynamics in a full-scale (568 m 3 ·d -1 ) microalgal system using 25 months of operational and microalgal community structure data. LSTM consistently achieved the highest prediction accuracy while effectively capturing system upset events. SHapley Additive exPlanations (SHAP) analysis provided interpretable insights into feature importance across both feature and temporal dimensions, with SHAP-guided feature reduction substantially reducing computational requirements without compromising predictive performance. Benchmarking against the mechanistic process model further identified the complementary application niches of mechanistic and data-driven approaches, providing practical guidance for model selection.

Related

  1. Electrode performance in hybrid electrocoagulation and biochar adsorption systems for olive mill wastewater treatment
    Ahmad Jamrah, Aseel Al-Khraisha, Khaldoun Shatanawi +2
    PLOS OnePeer-reviewedCC BYFull textPhysical chemistryEnvironmental science & pollutionPLOSEurope PMC
  2. Advanced lignin-derived carbon aerogel for efficient uranium removal from complex mining wastewater: adsorption performance, mechanism, and thermodynamics
    Mohammad Al-Shannag, Mohammad Alnaief, Ronaldo Antunes Funari +5
    HydrometallurgyPeer-reviewedCC BY5 citedInorganic & organometallic chemistryOpenAlex
  3. Eco-friendly method for the analysis of the soluble fraction of parabens leached from sludge of a hospital wastewater treatment plant—proof of concept
    S. Bedoya Patiño, J. P. Arrubla Vélez, J. M. Grisales Bedoya +5
    Analytical and Bioanalytical ChemistryPeer-reviewedCC BY 4.0Full textAnalytical chemistryPMC
  4. Conventional Ion-Exchange and Chelating Resins for Metal Recovery: A Critical Comparison of Selectivity, Dynamic Performance, Regeneration, and Process Integration
    Antonio Clareti Pereira,, PhD
    International Journal of Current Science Research and ReviewPeer-reviewedLicence not stated7 citedMaterials & polymersOpenAlex
  5. Process effect prediction method based on visual generative models
    Jianhong Zhao, Huamin Yang, Yi Sui +1
    Scientific ReportsPeer-reviewedCC BY 4.0Full textMaterials & polymersCrossref

Cite

BibTeXRIS
BibTeX
@misc{kim2026datadriven,
  title = {{Data-driven models for performance prediction in a microalgae-based wastewater treatment process}},
  author = {Ga-Yeong Kim and JooMan Kim and Jeremy S. Guest},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15009222/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009222/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Data-driven models for performance prediction in a microalgae-based wastewater treatment process
AU  - Ga-Yeong Kim
AU  - JooMan Kim
AU  - Jeremy S. Guest
PY  - 2026
DA  - 2026/09/22
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15009222/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009222/v1
AB  - Microalgae-based wastewater treatment has the potential to advance tertiary phosphorus recovery while producing biomass for resource valorization. The development of reliable near-term prediction models can help reduce system upsets, support proactive process control, and optimize resource utilization. In this study, Light Gradient-Boosting Machine (LGBM), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) models were comparatively evaluated for predicting effluent ammonium, nitrate, and orthophosphate concentrations and biomass dynamics in a full-scale (568 m 3 ·d -1 ) microalgal system using 25 months of operational and microalgal community structure data. LSTM consistently achieved the highest prediction accuracy while effectively capturing system upset events. SHapley Additive exPlanations (SHAP) analysis provided interpretable insights into feature importance across both feature and temporal dimensions, with SHAP-guided feature reduction substantially reducing computational requirements without compromising predictive performance. Benchmarking against the mechanistic process model further identified the complementary application niches of mechanistic and data-driven approaches, providing practical guidance for model selection.
ER  -