A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities
Abstract Reduced bacteria concentrations in wastewater is a key indicator of the efficacy of water resource recovery facilities (WRRFs). However, monitoring the presence of bacterial concentrations in real time at each stage of the WRRF is challenging as it requires taking and processing water sampl...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-82598-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846101253817368576 |
|---|---|
| author | Fahad Aljehani Ibrahima N’Doye Pei-Ying Hong Mohammad Khalil Monjed Taous-Meriem Laleg-Kirati |
| author_facet | Fahad Aljehani Ibrahima N’Doye Pei-Ying Hong Mohammad Khalil Monjed Taous-Meriem Laleg-Kirati |
| author_sort | Fahad Aljehani |
| collection | DOAJ |
| description | Abstract Reduced bacteria concentrations in wastewater is a key indicator of the efficacy of water resource recovery facilities (WRRFs). However, monitoring the presence of bacterial concentrations in real time at each stage of the WRRF is challenging as it requires taking and processing water samples offline. Although few studies have been proposed to predict bacterial concentrations using data-driven models, generalizing these models to unseen data from different WRRFs remains challenging. This paper proposes a calibration approach based on neural networks to adapt the optimal models across various WRRFs in Saudi Arabia for bacterial estimation at the influent and effluent stages. The calibration relies on the out-of-distribution (OOD) framework of the physiochemical water parameters (e.g., pH, COD, TDS, turbidity, conductivity) with a design threshold chosen based on the data distribution of the received unseen samples. We propose a calibration framework that continues updating the trained neural network model for accurate bacterial concentration estimation upon receiving new samples. We tested the effectiveness of the proposed calibration scheme on four WRRF datasets in Saudi Arabia, comparing the results with before and after calibration without the OOD. Before calibration model was based on a traditional and optimal neural network approach, typically considered the conventional method for building neural networks. After calibration without OOD, the model continued retraining without explicitly checking for OOD condition. The results showed that the proposed calibration framework of the selected baseline WRRF with the OOD scheme improved $$99.68\%$$ and $$56.00\%$$ of the worst-case influent bacteria concentration before calibration and after calibration without OOD, respectively. Similarly, the worst-case effluent bacteria concentration estimation was enhanced by $$99.37\%$$ before calibration and $$33.98\%$$ after calibration without the OOD. Our findings highlight the importance of integrating the calibration framework with neural network approaches to achieve model generalization. |
| format | Article |
| id | doaj-art-36e5a89bb9054f9e8680c43c1cd0f701 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-36e5a89bb9054f9e8680c43c1cd0f7012024-12-29T12:21:28ZengNature PortfolioScientific Reports2045-23222024-12-0114111410.1038/s41598-024-82598-yA calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilitiesFahad Aljehani0Ibrahima N’Doye1Pei-Ying Hong2Mohammad Khalil Monjed3Taous-Meriem Laleg-Kirati4Computer, Electrical and Mathematical Sciences and Engineering Division (CEMSE), King Abdullah University of Science and Technology (KAUST)Computer, Electrical and Mathematical Sciences and Engineering Division (CEMSE), King Abdullah University of Science and Technology (KAUST)Environmental Science and Engineering Program, Biological and Environmental Science and Engineering Division, King Abdullah University of Science and Technology (KAUST)Faculty of Science, Umm Al-Qura UniversityNational Institute for Research in Digital Science and Technology (INRIA)Abstract Reduced bacteria concentrations in wastewater is a key indicator of the efficacy of water resource recovery facilities (WRRFs). However, monitoring the presence of bacterial concentrations in real time at each stage of the WRRF is challenging as it requires taking and processing water samples offline. Although few studies have been proposed to predict bacterial concentrations using data-driven models, generalizing these models to unseen data from different WRRFs remains challenging. This paper proposes a calibration approach based on neural networks to adapt the optimal models across various WRRFs in Saudi Arabia for bacterial estimation at the influent and effluent stages. The calibration relies on the out-of-distribution (OOD) framework of the physiochemical water parameters (e.g., pH, COD, TDS, turbidity, conductivity) with a design threshold chosen based on the data distribution of the received unseen samples. We propose a calibration framework that continues updating the trained neural network model for accurate bacterial concentration estimation upon receiving new samples. We tested the effectiveness of the proposed calibration scheme on four WRRF datasets in Saudi Arabia, comparing the results with before and after calibration without the OOD. Before calibration model was based on a traditional and optimal neural network approach, typically considered the conventional method for building neural networks. After calibration without OOD, the model continued retraining without explicitly checking for OOD condition. The results showed that the proposed calibration framework of the selected baseline WRRF with the OOD scheme improved $$99.68\%$$ and $$56.00\%$$ of the worst-case influent bacteria concentration before calibration and after calibration without OOD, respectively. Similarly, the worst-case effluent bacteria concentration estimation was enhanced by $$99.37\%$$ before calibration and $$33.98\%$$ after calibration without the OOD. Our findings highlight the importance of integrating the calibration framework with neural network approaches to achieve model generalization.https://doi.org/10.1038/s41598-024-82598-yWater resource recovery facilitiesBacteria concentration sensingWasserstein generative adversarial network (WGAN)Out-of-distribution (OOD) generalizationCalibration of neural networks |
| spellingShingle | Fahad Aljehani Ibrahima N’Doye Pei-Ying Hong Mohammad Khalil Monjed Taous-Meriem Laleg-Kirati A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities Scientific Reports Water resource recovery facilities Bacteria concentration sensing Wasserstein generative adversarial network (WGAN) Out-of-distribution (OOD) generalization Calibration of neural networks |
| title | A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| title_full | A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| title_fullStr | A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| title_full_unstemmed | A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| title_short | A calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| title_sort | calibration framework toward model generalization for bacteria concentration estimation in water resource recovery facilities |
| topic | Water resource recovery facilities Bacteria concentration sensing Wasserstein generative adversarial network (WGAN) Out-of-distribution (OOD) generalization Calibration of neural networks |
| url | https://doi.org/10.1038/s41598-024-82598-y |
| work_keys_str_mv | AT fahadaljehani acalibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT ibrahimandoye acalibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT peiyinghong acalibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT mohammadkhalilmonjed acalibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT taousmeriemlalegkirati acalibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT fahadaljehani calibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT ibrahimandoye calibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT peiyinghong calibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT mohammadkhalilmonjed calibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities AT taousmeriemlalegkirati calibrationframeworktowardmodelgeneralizationforbacteriaconcentrationestimationinwaterresourcerecoveryfacilities |