Current machine learning (ML) efforts for predicting hydrogen yield in dark fermentation are constrained by limited sample sizes and distributional skewness, yielding unstable models. These data characteristics fundamentally restrict generalization and hinder the optimization of process conditions. In this study, a generative adversarial network (GAN)-inspired strategy was developed to augment an initial dataset of 210 dark fermentation samples to 1050 synthetic instances, significantly enhancing data distribution normality and coverage. Across nine ML algorithms, the Histogram-based Gradient Boosting (HGB) model performed best on the test dataset ( R2 ≈ 0.95; RMSE < 0.06; MAE < 0.05). SHAP and accumulated local effects (ALE) analyses indicated that butyrate, biomass, and Ni positively influenced hydrogen yield, whereas elevated COD, ethanol, and longer hydraulic retention time (HRT) reduced it. Two-dimensional ALE plots further identified the optimal operating conditions for dark fermentation (Fe/Ni ratio ≈ 1:3; HRT of 4–5 h; pH ≈ 4.9; and COD < 25 g L −1 ). A Python-based graphical user interface (GUI) integrating the HGB model was developed for practical hydrogen yield prediction and process diagnostics. This study demonstrates that combining GAN-inspired data with gradient boosting models can enhance both prediction accuracy and process control in biohydrogen production from wastewater.
目前用于预测暗发酵产氢的机器学习（ML）努力受到样本量和分布偏度的限制，产生不稳定的模型。这些数据特征从根本上限制了泛化，阻碍了工艺条件的优化。在本研究中，开发了一种生成对抗网络（GAN）启发的策略，将210个暗发酵样本的初始数据集扩展到1050个合成实例，显著增强了数据分布的正态性和覆盖率。在9种ML算法中，基于直方图的梯度增强（HGB）模型在测试数据集上表现最好（r2≈0.95;RMSE < 0.06; MAE < 0.05）。SHAP和累积局部效应（ALE）分析表明，丁酸盐、生物量和Ni对产氢率有积极影响，而COD、乙醇和较长的水力滞留时间（HRT）则会降低产氢率。二维ALE图进一步确定了暗发酵的最佳操作条件（Fe/Ni比≈1:3，HRT为4-5 h， pH≈4.9,COD < 25 g L−1）。结合HGB模型，开发了一个基于python的图形用户界面（GUI），用于实际产氢量预测和过程诊断。该研究表明，将gan启发的数据与梯度增强模型相结合，可以提高废水生物制氢的预测精度和过程控制。
{"title":"Augmented machine learning with limited data for hydrogen yield prediction in wastewater dark fermentation","authors":"Chong Liu, Fayong Li, Pengyan Zhang, Paramasivan Balasubramanian","doi":"10.1038/s41545-025-00529-4","DOIUrl":"https://doi.org/10.1038/s41545-025-00529-4","url":null,"abstract":"Current machine learning (ML) efforts for predicting hydrogen yield in dark fermentation are constrained by limited sample sizes and distributional skewness, yielding unstable models. These data characteristics fundamentally restrict generalization and hinder the optimization of process conditions. In this study, a generative adversarial network (GAN)-inspired strategy was developed to augment an initial dataset of 210 dark fermentation samples to 1050 synthetic instances, significantly enhancing data distribution normality and coverage. Across nine ML algorithms, the Histogram-based Gradient Boosting (HGB) model performed best on the test dataset ( <jats:italic>R</jats:italic> <jats:sup>2</jats:sup> ≈ 0.95; RMSE < 0.06; MAE < 0.05). SHAP and accumulated local effects (ALE) analyses indicated that butyrate, biomass, and Ni positively influenced hydrogen yield, whereas elevated COD, ethanol, and longer hydraulic retention time (HRT) reduced it. Two-dimensional ALE plots further identified the optimal operating conditions for dark fermentation (Fe/Ni ratio ≈ 1:3; HRT of 4–5 h; pH ≈ 4.9; and COD < 25 g L <jats:sup>−</jats:sup> <jats:sup>1</jats:sup> ). A Python-based graphical user interface (GUI) integrating the HGB model was developed for practical hydrogen yield prediction and process diagnostics. This study demonstrates that combining GAN-inspired data with gradient boosting models can enhance both prediction accuracy and process control in biohydrogen production from wastewater.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"67 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145611187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1038/s41545-025-00530-x
Yong-Uk Shin, Dongwoo Kim, Sung Il Yu, Hyokwan Bae, Am Jang
{"title":"Smart control of oxychlorine species using reinforcement learning in saline electrochemical oxidation","authors":"Yong-Uk Shin, Dongwoo Kim, Sung Il Yu, Hyokwan Bae, Am Jang","doi":"10.1038/s41545-025-00530-x","DOIUrl":"https://doi.org/10.1038/s41545-025-00530-x","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"175 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145611264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-23DOI: 10.1038/s41545-025-00531-w
Charles Balkenbusch, Judith Glienke, Yuhao Wu, Keenan Munno, Michael Jung, Husein Almuhtaram, Robert C. Andrews
{"title":"Microplastic removal across ten drinking water treatment facilities and distribution systems","authors":"Charles Balkenbusch, Judith Glienke, Yuhao Wu, Keenan Munno, Michael Jung, Husein Almuhtaram, Robert C. Andrews","doi":"10.1038/s41545-025-00531-w","DOIUrl":"https://doi.org/10.1038/s41545-025-00531-w","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"138 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145575338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-18DOI: 10.1038/s41545-025-00524-9
Xianglan Jiao, Tao Xia, Lingzhi Zhang, Lei Shi, Zhimin Ao, Xuede Li, Jie Li
{"title":"Phosphorus adsorption in paddy water by immobilized Ce-MOFs: performance, mechanism analysis, and dynamic adsorption","authors":"Xianglan Jiao, Tao Xia, Lingzhi Zhang, Lei Shi, Zhimin Ao, Xuede Li, Jie Li","doi":"10.1038/s41545-025-00524-9","DOIUrl":"https://doi.org/10.1038/s41545-025-00524-9","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"1 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145545481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-18DOI: 10.1038/s41545-025-00526-7
Milad Mohsenzadeh, Shuqi Xu, Osman Shamet, Juan F. Torres
{"title":"Scalable brine treatment using 3D-printed multichannel thermodiffusion","authors":"Milad Mohsenzadeh, Shuqi Xu, Osman Shamet, Juan F. Torres","doi":"10.1038/s41545-025-00526-7","DOIUrl":"https://doi.org/10.1038/s41545-025-00526-7","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"15 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145536155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-18DOI: 10.1038/s41545-025-00528-5
Christopher Jackson, Shuqi Xu, Juan F. Torres
{"title":"Techno-economic analysis of multichannel thermodiffusion for desalination and brine concentration","authors":"Christopher Jackson, Shuqi Xu, Juan F. Torres","doi":"10.1038/s41545-025-00528-5","DOIUrl":"https://doi.org/10.1038/s41545-025-00528-5","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"30 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145536327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Remote sensing monitoring of small-lake eutrophication faces challenges such as sparse data, insufficient synergy of multi-source data, and limited model generalization performance. Hence, this study developed a scenario-aware modeling framework for the trophic level index (TLI) by integrating multi-source imagery data from Sentinel-2, GF-1, HJ-2, and PlanetScope, using Dongqian Lake in Zhejiang Province, China as the case study. The cross-sensor prediction accuracy was evaluated using algorithms such as CatBoost Regression (CBR), XGBoost Regression (XGBR), TabPFN Regression (TPFNR), and Linear Regression (LR). Meanwhile, the influence of input features was quantified by SHapley Additive exPlanations (SHAP). The main results found that : (1) Overall annual mean values of total nitrogen/total phosphorus ratio (TN/TP) and TLI were 22.13 and 37.36 ± 4.99, respectively, indicating a mesotrophic and phosphorus-limited state in Dongqian Lake. (2) TLI exhibited the strongest correlation with water color and algal spectral indices, including Normalized Difference Water Index (NDWI), Normalized Green–Red Difference Index (NGRDI), and Blue–Green Ratio (BGR). (3) CBR demonstrated the strongest cross-sensor generalization capability across different imagery, with only minor variations in prediction accuracy (ΔR ≈ 0.07–0.15). Feature attribution analysis identified NDWI, NGRDI, and BGR as primary contributing features for the CBR model. (4) Integrating high-frequency multi-source remote sensing imagery with 27 field surveys achieved seamless monitoring of the TLI. The spatial distribution of TLI showed distinct seasonal variations, with higher values observed in nearshore areas and lower values in the lake center. TLI values were relatively low in spring, but surged sharply and remained elevated in summer. This study provided a reference basis for detailed remote sensing monitoring and management of eutrophication in small lakes.
{"title":"Resolving inherent constraints in eutrophication monitoring of small lakes using multi-source satellites and machine learning","authors":"Wei Si, Zhixiong Chen, Chi Yung Jim, Mou Leong Tan, Dong Liu, Yue Yao, Lifei Wei, Shangshang Xu, Fei Zhang","doi":"10.1038/s41545-025-00525-8","DOIUrl":"https://doi.org/10.1038/s41545-025-00525-8","url":null,"abstract":"Remote sensing monitoring of small-lake eutrophication faces challenges such as sparse data, insufficient synergy of multi-source data, and limited model generalization performance. Hence, this study developed a scenario-aware modeling framework for the trophic level index (TLI) by integrating multi-source imagery data from Sentinel-2, GF-1, HJ-2, and PlanetScope, using Dongqian Lake in Zhejiang Province, China as the case study. The cross-sensor prediction accuracy was evaluated using algorithms such as CatBoost Regression (CBR), XGBoost Regression (XGBR), TabPFN Regression (TPFNR), and Linear Regression (LR). Meanwhile, the influence of input features was quantified by SHapley Additive exPlanations (SHAP). The main results found that : (1) Overall annual mean values of total nitrogen/total phosphorus ratio (TN/TP) and TLI were 22.13 and 37.36 ± 4.99, respectively, indicating a mesotrophic and phosphorus-limited state in Dongqian Lake. (2) TLI exhibited the strongest correlation with water color and algal spectral indices, including Normalized Difference Water Index (NDWI), Normalized Green–Red Difference Index (NGRDI), and Blue–Green Ratio (BGR). (3) CBR demonstrated the strongest cross-sensor generalization capability across different imagery, with only minor variations in prediction accuracy (ΔR ≈ 0.07–0.15). Feature attribution analysis identified NDWI, NGRDI, and BGR as primary contributing features for the CBR model. (4) Integrating high-frequency multi-source remote sensing imagery with 27 field surveys achieved seamless monitoring of the TLI. The spatial distribution of TLI showed distinct seasonal variations, with higher values observed in nearshore areas and lower values in the lake center. TLI values were relatively low in spring, but surged sharply and remained elevated in summer. This study provided a reference basis for detailed remote sensing monitoring and management of eutrophication in small lakes.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"32 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145536154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-07DOI: 10.1038/s41545-025-00527-6
Leonardo E. Navarrete-Cevallos, Ronald Vargas, Patricio J. Espinoza-Montero
{"title":"Publisher Correction: Fundamentals and environmental applications of bismuth vanadate through photoelectrocatalysis","authors":"Leonardo E. Navarrete-Cevallos, Ronald Vargas, Patricio J. Espinoza-Montero","doi":"10.1038/s41545-025-00527-6","DOIUrl":"https://doi.org/10.1038/s41545-025-00527-6","url":null,"abstract":"","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"93 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study presents a life cycle assessment (LCA) of a scaled-up photoelectrocatalytic (PEC) oxidation system for wastewater treatment, modelled using computational fluid dynamics (CFD). The system used a BiVO 4 /TiO 2 -GO photoanode for solar-driven degradation of micropollutants. The LCA assesses energy use, resource demand, and emissions to evaluate the system’s sustainability in line with EU wastewater regulations. Compared to a full-scale ozonation plant in the Netherlands, the PEC system shows superior environmental performance during operation and end-of-life phases, despite higher construction impacts. Solar energy use and potential material reuse drive these advantages. A comparison with theoretical pilot-scale oxidation technologies from literature adds depth, though the study acknowledges limitations such as micropollutant variability and wastewater complexity. Overall, the findings highlight PEC oxidation’s promise as a sustainable and effective approach for micropollutant removal in water treatment.
{"title":"Sustainable advanced wastewater treatment via photoelectrocatalytic oxidation: insights from life cycle assessment","authors":"Gema Amaya Santos, Agha Zeeshan Ali, Paola Lettieri","doi":"10.1038/s41545-025-00522-x","DOIUrl":"https://doi.org/10.1038/s41545-025-00522-x","url":null,"abstract":"This study presents a life cycle assessment (LCA) of a scaled-up photoelectrocatalytic (PEC) oxidation system for wastewater treatment, modelled using computational fluid dynamics (CFD). The system used a BiVO <jats:sub>4</jats:sub> /TiO <jats:sub>2</jats:sub> -GO photoanode for solar-driven degradation of micropollutants. The LCA assesses energy use, resource demand, and emissions to evaluate the system’s sustainability in line with EU wastewater regulations. Compared to a full-scale ozonation plant in the Netherlands, the PEC system shows superior environmental performance during operation and end-of-life phases, despite higher construction impacts. Solar energy use and potential material reuse drive these advantages. A comparison with theoretical pilot-scale oxidation technologies from literature adds depth, though the study acknowledges limitations such as micropollutant variability and wastewater complexity. Overall, the findings highlight PEC oxidation’s promise as a sustainable and effective approach for micropollutant removal in water treatment.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"82 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145411741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-31DOI: 10.1038/s41545-025-00523-w
Carley S. Truyens, David M. Berendes, Molly E. Cantrell, Alexandra L. Kossik, Kerrigan M. McCarthy, Anna S. Mehrotra, Jennifer L. Murphy, Sudhir Pillay, Suraja J. Raj, Maya S. Ramaswamy, Habib Yakubu, Rochelle H. Holm
Wastewater and environmental surveillance is a valuable tool for early warning, detection, and response to emerging public health threats, with the added ability to inform data gaps across several Sustainable Development Goals. Drawing from our experiences in Bangladesh, Ghana, Malawi, and South Africa, we call to action this often unmentioned link through critical applied research questions and engagement in peer-to-peer learning and global Communities of Practice.
{"title":"Advancing wastewater and environmental surveillance in LMICs for public health response and SDG data gaps","authors":"Carley S. Truyens, David M. Berendes, Molly E. Cantrell, Alexandra L. Kossik, Kerrigan M. McCarthy, Anna S. Mehrotra, Jennifer L. Murphy, Sudhir Pillay, Suraja J. Raj, Maya S. Ramaswamy, Habib Yakubu, Rochelle H. Holm","doi":"10.1038/s41545-025-00523-w","DOIUrl":"https://doi.org/10.1038/s41545-025-00523-w","url":null,"abstract":"Wastewater and environmental surveillance is a valuable tool for early warning, detection, and response to emerging public health threats, with the added ability to inform data gaps across several Sustainable Development Goals. Drawing from our experiences in Bangladesh, Ghana, Malawi, and South Africa, we call to action this often unmentioned link through critical applied research questions and engagement in peer-to-peer learning and global Communities of Practice.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"1 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145411773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：