@article{khaled_explainable_2026a,

title = {Explainable Fuzzy GNNs for Leak Detection in Water Distribution Networks},

author = {Q. Khaled and P. De Marinis and M. Louati and D. Ferras and L. Genga and U. Kaymak },

year  = {2026},

date = {2026-07-29},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{khaled_explainable_2026,

title = {Explainable Uncertainty Quantification for Wastewater Treatment Energy Prediction via Interval Type-2 Neuro-Fuzzy System},

author = {Q. Khaled and B. Mallak and U. Kaymak and L. Genga},

year  = {2026},

date = {2026-07-29},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{khaled_predictive_2026,

title = {Predictive Maintenance for Ultrafiltration Membranes Using Explainable Similarity-Based Prognostics},

author = {Q. Khaled and L. Genga and U. Kaymak},

year  = {2026},

date = {2026-07-29},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{güven_exploring_2026,

title = {Exploring Sustainable Water Management and Energy Transition In Curaçao: ILUSTRE Lab Consortium Visit report},

author = {Ç. Güven and U. Kaymak and Q. Khaled and M. Latifi and I. Ahmed S. and K. Mwandingi and M. Noorman and P. Ugochukwu Orji and R. Salessa and C. Zucca},

url = {https://research.tilburguniversity.edu/en/publications/exploring-sustainable-water-management-and-energy-transition-in-c/?_gl=1*1r3bpb8*_gcl_aw*R0NMLjE3NjkwNTAxODguQ2p3S0NBaUFqOExMQmhBa0Vpd0FKamJZNzhiRDRyWjloZmdKV1AyYm9sd2RYcTNlR1hTbGx5NDIzUWl5WkZFdVBjY3BJNzMxSGtrX0xSb0NyU1FRQXZEX0J3RQ..*_gcl_au*ODk3MjE0MjA1LjE3NjYzOTI4MTk.*FPAU*ODk3MjE0MjA1LjE3NjYzOTI4MTk.},

year  = {2026},

date = {2026-07-29},

urldate = {2026-07-29},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{orji_gridaware_2025,

title = {Grid-Aware Spatio-Temporal Graph Neural Networks for Multi-Horizon Load Forecasting},

author = {U. Orji and Ç. Güven and D. Stowell},

year  = {2025},

date = {2025-07-30},

volume = {5},

number = {3},

pages = {52–65},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{latifi_federated_2025,

title = {A federated game-theoretic strategy for three-phase voltage unbalance mitigation via dynamic pricing and demand response},

author = {M. Latifi and C. Zucca and M. C. Willemsen},

url = {https://www.sciencedirect.com/science/article/pii/S2352467725004412},

doi = {https://doi.org/10.1016/j.segan.2025.102059},

issn = {2352-4677},

year  = {2025},

date = {2025-07-30},

volume = {44},

pages = {102059},

abstract = {Request PDF textbar On Jan 1, 2025, Milad Latifi and others published A Federated Game-Theoretic Strategy for Three-Phase Voltage Unbalance Mitigation Via Dynamic Pricing and Demand Response textbar Find, read and cite all the research you need on ResearchGate},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{orji_enhanced_2025,

title = {Enhanced load forecasting with gat-lstm: Leveraging grid and temporal features},

author = {U. Orji and Ç. Güven and D. Stowell},

year  = {2025},

date = {2025-07-29},

abstract = {Accurate power load forecasting is essential for the efficient operation and planning of electrical grids, particularly given the increased variability and complexity introduced by renewable energy sources. This paper introduces GAT-LSTM, a hybrid model that combines Graph Attention Networks (GAT) and Long Short-Term Memory (LSTM) networks. A key innovation of the model is the incorporation of edge attributes, such as line capacities and efficiencies, into the attention mechanism, enabling it to dynamically capture spatial relationships grounded in grid-specific physical and operational constraints. Additionally, by employing an early fusion of spatial graph embeddings and temporal sequence features, the model effectively learns and predicts complex interactions between spatial dependencies and temporal patterns, providing a realistic representation of the dynamics of power grids. Experimental evaluations on the Brazilian Electricity System dataset demonstrate that the GAT-LSTM model significantly outperforms state-of-the-art models, achieving reductions of 21. 8% in MAE, 15. 9% in RMSE and 20. 2% in MAPE. These results underscore the robustness and adaptability of the GAT-LSTM model, establishing it as a powerful tool for applications in grid management and energy planning.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{khaled_end–end_2025,

title = {An End-to-End Framework for AI Integration in Desalination Systems: 20th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU2024},

author = {Q. Khaled and U. Kaymak and L. Genga},

editor = {M. J. Lesot, S. Vieira, M. Reformat, F. Batista, J. P. Carvalho, B. Bouchon-Menier, R. R. Yager},

doi = {10.5281/zenodo.15149423},

year  = {2025},

date = {2025-04-01},

urldate = {2025-04-01},

journal = {IPMU2024 Lisboa - Short Paper Proceedings},

pages = {39–43},

abstract = {In the context of population growth, urbanization, and industrial expansion, water scarcity emerges as a significant concern, with projections indicating that around two billion individuals may face this challenge by 2050. Hence, the increased pressure on existing water resources calls for new water supply solutions in light of the growing demand. Desalination emerges as a promising alternative solution, particularly in regions confronting limited water resources. The sector has experienced remarkable growth, witnessing a 41% capacity increase over the past decade, with projections hinting at a twofold expansion by 2030. Such expansion requires integrating cutting-edge modeling techniques to ensure efficacy and cost-effectiveness. Artificial intelligence (AI) shows potential to revolutionize desalination and water treatment practices, yet its implementation remains limited. Delayed integration is believed to stem from the lack of trust among domain experts, knowledge gaps between water professionals and data scientists, and untapped potential within the field. This paper proposes The Integrated System Perspective for AI-based Desalination (ISP); an End-to-End Framework for AI in desalination. ISP-AID facilitates identifying AI applications across various project stages, from design to maintenance, uncovering opportunities for cost reduction and efficiency improvement. It adopts a structured data science perspective, integrating the Cross-Industry Standard Process for Data Mining (CRISP-DM) to guide AI algorithm selection and deployment. Spanning project cycle, process design, and data science levels, the framework aims to instill trust, foster collaborative problem understanding, and highlight untapped potential. This positions domain experts to actively develop data-driven solutions and enhancing confidence in innovative methodologies. By facilitating collaboration and exploring AI applications, the framework could expedite adopting efficient desalination solutions, thereby addressing global water scarcity challenges.},

note = {Publisher: Zenodo},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{khaled_semi-supervised_2025,

title = {A semi-supervised graph-based approach for anomaly detection in river monitoring stations: NCR Days 2025 - Crossing Boundaries},

author = {Q. Khaled and U. Kaymak and L. Genga},

editor = {V. Chavarrias, A. M. Hoek},

year  = {2025},

date = {2025-01-01},

urldate = {2025-01-01},

journal = {Crossing boundaries},

volume = {57-2025},

pages = {28–29},

abstract = {Water quality monitoring is crucial for ensuring safe drinking water and protecting aquatic ecosystems. Traditional methods, like periodic sampling, often fail to detect sudden pollution events, delaying responses that could protect public health and the environment Zhu et al. (2022). Anomalies, such as nutrient spikes or oxygen drops, can signal environmental issues like pollution or runoff, necessitating immediate detection. This gap highlights the need for advanced techniques capable of realtime data analysis. Machine learning, particularly graph neural networks (GNNs), has shown promise in enhancing real-time water quality monitoring. GNNs are effective at modeling the spatial connections between monitoring stations, which helps in predicting water quality more accurately Li et al. (2024) Yan and Wang (2024). This is especially useful for understanding how pollution spreads across a network. However, a significant hurdle is the scarcity of labeled data for training GNNs, as anomalies in water quality are rare and often not well-documented Buchhorn et al. (2024). This makes supervised learning challenging, pushing researchers toward novel methods to detect unusual events without extensive labeled datasets.},

note = {Publisher: Netherlands Centre for River Studies},

keywords = {},

pubstate = {published},

tppubtype = {article}

}