Pub Date : 2025-12-31DOI: 10.1038/s41545-025-00546-3
Claude Kiki, Qian Sun
Bacterial granular sludge (BGS) efficiently removes antibiotics but acts as a hotspot for the enrichment of antibiotic resistance genes (ARGs). To decouple removal from resistance, we integrated Microcystis aeruginosa with BGS to form a cyanobacterial–bacterial granular sludge (CBGS) system. During 120-day operation under multi-class antibiotic stress (0–500 µg L⁻¹), CBGS demonstrated enhanced structural stability, pollutant removal, and a higher antibiotic elimination capacity (93.71–671.93 vs. 93.38–499.63 µg L⁻¹ d⁻¹ in BGS). Simultaneously, the system achieved suppression of endogenous microcystin-LR (a cyanotoxin variant containing Leucine and Arginine) release. Metagenomic, network, and transformation product analyses revealed that cyanobacterial integration induced a functional decoupling within the community, promoting oxidative biodegradation pathways and reducing selective pressure on ARG hosts. Mobility gene analysis further indicated restricted horizontal gene transfer, limiting ARG exchange between biomass and effluent. Consequently, ARG abundance decreased in effluents for 13 of 20 types in CBGS but increased for 16 of 20 in BGS. This study deepens our understanding of how microbial consortia can be engineered to separate antibiotic degradation from resistance propagation, offering a promising biologically contained strategy to mitigate resistance risks in wastewater treatment.
细菌颗粒污泥（BGS）能有效去除抗生素，但也是抗生素耐药基因（ARGs）富集的热点。为了将去除与耐药性分离，我们将铜绿微囊藻与BGS结合，形成蓝藻-细菌颗粒污泥（CBGS）系统。在120天的多种抗生素压力下（0-500µg L -⁻¹），CBGS表现出更强的结构稳定性，去除污染物和更高的抗生素消除能力（93.71-671.93µg L -⁻¹vs. BGS的93.38-499.63µg L -⁻¹）。同时，该系统还抑制了内源性微囊藻毒素lr（一种含有亮氨酸和精氨酸的蓝藻毒素变体）的释放。宏基因组、网络和转化产物分析表明，蓝藻整合诱导了群落内的功能解耦，促进了氧化生物降解途径，减少了对ARG宿主的选择压力。移动性基因分析进一步表明，水平基因转移受限，限制了生物质与出水间ARG的交换。因此，在CBGS的20种类型中，有13种出水中ARG丰度下降，而在BGS的20种类型中，有16种出水中ARG丰度增加。这项研究加深了我们对微生物联合体如何被设计来分离抗生素降解和耐药性繁殖的理解，为减轻废水处理中的耐药性风险提供了一种有前途的生物控制策略。
{"title":"Decoupling antibiotic degradation from resistance development: photogranules act as a biocontainment for ARGs in wastewater","authors":"Claude Kiki, Qian Sun","doi":"10.1038/s41545-025-00546-3","DOIUrl":"https://doi.org/10.1038/s41545-025-00546-3","url":null,"abstract":"Bacterial granular sludge (BGS) efficiently removes antibiotics but acts as a hotspot for the enrichment of antibiotic resistance genes (ARGs). To decouple removal from resistance, we integrated Microcystis aeruginosa with BGS to form a cyanobacterial–bacterial granular sludge (CBGS) system. During 120-day operation under multi-class antibiotic stress (0–500 µg L⁻¹), CBGS demonstrated enhanced structural stability, pollutant removal, and a higher antibiotic elimination capacity (93.71–671.93 vs. 93.38–499.63 µg L⁻¹ d⁻¹ in BGS). Simultaneously, the system achieved suppression of endogenous microcystin-LR (a cyanotoxin variant containing Leucine and Arginine) release. Metagenomic, network, and transformation product analyses revealed that cyanobacterial integration induced a functional decoupling within the community, promoting oxidative biodegradation pathways and reducing selective pressure on ARG hosts. Mobility gene analysis further indicated restricted horizontal gene transfer, limiting ARG exchange between biomass and effluent. Consequently, ARG abundance decreased in effluents for 13 of 20 types in CBGS but increased for 16 of 20 in BGS. This study deepens our understanding of how microbial consortia can be engineered to separate antibiotic degradation from resistance propagation, offering a promising biologically contained strategy to mitigate resistance risks in wastewater treatment.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"15 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145894499","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 : 2025-12-30DOI: 10.1038/s41545-025-00547-2
Changjian Ma, Enkai Cao, Bowen Li, Ning Shi, Zeqiang Sun, Yan Li, Peng Hou, Yang Xiao
Biogas slurry, a liquid by-product of anaerobic digestion, is increasingly reused in agriculture, but its low phosphorus content often requires external supplementation. When phosphorus is introduced into slurry fertigation, it interacts with Ca²⁺, Mg²⁺, bicarbonates, and organic matter, creating conditions that can accelerate pipeline clogging. In this study, we combined a 60-day accelerated pipeline experiment with mineralogical analysis, scanning electron microscopy, and high-throughput sequencing to examine how wastewater–phosphorus coupling influences fouling. Slurry alone caused little short-term clogging, but the addition of phosphorus led to sharp increases in fouling mass (up to 130%) and flow reduction (up to 90%). Mineralogical analyses identified secondary phosphate precipitates such as brushite, baricite, and apatite, while microbial community profiling showed greater diversity, persistence, and biofilm-forming capacity under phosphorus conditions. Correlation analysis and structural equation modeling demonstrated that precipitates provided scaffolds for biofilm growth, particulates enhanced microbial attachment, and biofilms linked physical and chemical processes to hydraulic decline. These results show that pipeline fouling under wastewater–phosphorus coupling arises from the synergy of biofilms, minerals, and particulates, and they offer mechanistic guidance for designing fouling control systems that support sustainable reuse of livestock wastewater.
{"title":"Wastewater–phosphorus coupling accelerates biofilm–mineral–particulate interactions in irrigation pipelines","authors":"Changjian Ma, Enkai Cao, Bowen Li, Ning Shi, Zeqiang Sun, Yan Li, Peng Hou, Yang Xiao","doi":"10.1038/s41545-025-00547-2","DOIUrl":"https://doi.org/10.1038/s41545-025-00547-2","url":null,"abstract":"Biogas slurry, a liquid by-product of anaerobic digestion, is increasingly reused in agriculture, but its low phosphorus content often requires external supplementation. When phosphorus is introduced into slurry fertigation, it interacts with Ca²⁺, Mg²⁺, bicarbonates, and organic matter, creating conditions that can accelerate pipeline clogging. In this study, we combined a 60-day accelerated pipeline experiment with mineralogical analysis, scanning electron microscopy, and high-throughput sequencing to examine how wastewater–phosphorus coupling influences fouling. Slurry alone caused little short-term clogging, but the addition of phosphorus led to sharp increases in fouling mass (up to 130%) and flow reduction (up to 90%). Mineralogical analyses identified secondary phosphate precipitates such as brushite, baricite, and apatite, while microbial community profiling showed greater diversity, persistence, and biofilm-forming capacity under phosphorus conditions. Correlation analysis and structural equation modeling demonstrated that precipitates provided scaffolds for biofilm growth, particulates enhanced microbial attachment, and biofilms linked physical and chemical processes to hydraulic decline. These results show that pipeline fouling under wastewater–phosphorus coupling arises from the synergy of biofilms, minerals, and particulates, and they offer mechanistic guidance for designing fouling control systems that support sustainable reuse of livestock wastewater.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"8 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145894503","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 : 2025-12-23DOI: 10.1038/s41545-025-00542-7
Sung-Hyo Jung, Sung-Hak Hong, Youn-Jun Lee, Seong-Jik Park, Eun Hea Jho, Chang-Gu Lee
Azo is a synthetic organic dye that has attracted considerable attention because of its recalcitrance to degradation and toxicity. An upgraded ozonation process must be developed that integrates decolorization, mineralization, and toxicity reduction to manage the residual azo dye in the effluent from the dyeing industry and reduce its associated aquatic environmental risks. In this study, an ozonation system using Meretrix lusoria (ML) shell-waste-derived biomass (ML800) that uses simple calcination was developed. The ML800/O3 system almost completely decolorized ( > 99.0%) and highly mineralized (53.6 ± 1.7%) Congo red (CR) during (CR = 100 mg∙L–1, ML800 = 0.5 g ∙ L–1), surpassing the performance of the other tested systems (single ozonation and single ML800). Moreover, the ML800/O3 system reduced the acute toxicity of CR to the bacterium Aliivibrio fischeri, whereas single ozonation showed temporarily increased the toxicity of CR. The FE-SEM/EDS, FTIR, and XRD analyses verified that Ca(OH)2 was the main calcium species in ML800, which catalyzed the decomposition of O3 into highly reactive •OH. The system was successfully applied to various azo dyes and was robust with water matrix constituents. These findings highlight the potential of marine shell waste for use as a sustainable and ecofriendly additive for ozonation, increasing azo dye removal from wastewater in practical applications.
{"title":"Enhancing ozonation using Meretrix lusoria shell waste biomass: sustainable decontamination of azo dye wastewater via decolorization, mineralization, and detoxification","authors":"Sung-Hyo Jung, Sung-Hak Hong, Youn-Jun Lee, Seong-Jik Park, Eun Hea Jho, Chang-Gu Lee","doi":"10.1038/s41545-025-00542-7","DOIUrl":"https://doi.org/10.1038/s41545-025-00542-7","url":null,"abstract":"Azo is a synthetic organic dye that has attracted considerable attention because of its recalcitrance to degradation and toxicity. An upgraded ozonation process must be developed that integrates decolorization, mineralization, and toxicity reduction to manage the residual azo dye in the effluent from the dyeing industry and reduce its associated aquatic environmental risks. In this study, an ozonation system using Meretrix lusoria (ML) shell-waste-derived biomass (ML800) that uses simple calcination was developed. The ML800/O3 system almost completely decolorized ( > 99.0%) and highly mineralized (53.6 ± 1.7%) Congo red (CR) during (CR = 100 mg∙L–1, ML800 = 0.5 g ∙ L–1), surpassing the performance of the other tested systems (single ozonation and single ML800). Moreover, the ML800/O3 system reduced the acute toxicity of CR to the bacterium Aliivibrio fischeri, whereas single ozonation showed temporarily increased the toxicity of CR. The FE-SEM/EDS, FTIR, and XRD analyses verified that Ca(OH)2 was the main calcium species in ML800, which catalyzed the decomposition of O3 into highly reactive •OH. The system was successfully applied to various azo dyes and was robust with water matrix constituents. These findings highlight the potential of marine shell waste for use as a sustainable and ecofriendly additive for ozonation, increasing azo dye removal from wastewater in practical applications.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"28 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145808170","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 : 2025-12-19DOI: 10.1038/s41545-025-00540-9
Tianxiang Wang, Guangyu Su, Simiao Wang, Yuxin Fan, Jin-Yu Terence Yang, Xinhan Xu, Zixiong Wang, Jian Wu, Qiuhua Liang, Yin Su, Jingjing Zhan, Lifen Liu, George Arhonditsis
The diminution of the benefits of domestic pollution control by greenhouse gas (GHG) emissions has received considerable attention. Emission factors related to the construction and operation of wastewater treatment systems have been well characterized in urban settings but far less so in rural areas. To address this gap, we developed an integrative modeling framework that quantifies the entire chain of rural domestic pollution processes together with the associated GHG emissions. Our analysis suggests that the control of China’s rural domestic pollution has realized a threefold increase over the past decade, resulting in a decline of carbon (C), nitrogen (N), and phosphorus (P) discharge to surface waters by 1158 Gg, 316 Gg, and 43 Gg, respectively. However, GHG emissions have also discernibly increased from 26.7 Tg to 31.4 Tg. Even though over 70% of China’s rural domestic pollution is still being discharged untreated, GHG emissions from wastewater treatment systems have become prevalent and currently account for more than 60% of total GHG emissions from rural areas. Considering the on-going construction of numerous new wastewater treatment systems in rural areas, enhancing wastewater treatment capacity, strengthening resource recovery, optimizing dietary patterns of the public, and promoting the use of clean energy are recommended to balance the trade-offs between environmental pollution abatement and climate change mitigation.
{"title":"China’s enhanced wastewater treatment capacity may accelerate greenhouse gas emissions from rural domestic pollution","authors":"Tianxiang Wang, Guangyu Su, Simiao Wang, Yuxin Fan, Jin-Yu Terence Yang, Xinhan Xu, Zixiong Wang, Jian Wu, Qiuhua Liang, Yin Su, Jingjing Zhan, Lifen Liu, George Arhonditsis","doi":"10.1038/s41545-025-00540-9","DOIUrl":"https://doi.org/10.1038/s41545-025-00540-9","url":null,"abstract":"The diminution of the benefits of domestic pollution control by greenhouse gas (GHG) emissions has received considerable attention. Emission factors related to the construction and operation of wastewater treatment systems have been well characterized in urban settings but far less so in rural areas. To address this gap, we developed an integrative modeling framework that quantifies the entire chain of rural domestic pollution processes together with the associated GHG emissions. Our analysis suggests that the control of China’s rural domestic pollution has realized a threefold increase over the past decade, resulting in a decline of carbon (C), nitrogen (N), and phosphorus (P) discharge to surface waters by 1158 Gg, 316 Gg, and 43 Gg, respectively. However, GHG emissions have also discernibly increased from 26.7 Tg to 31.4 Tg. Even though over 70% of China’s rural domestic pollution is still being discharged untreated, GHG emissions from wastewater treatment systems have become prevalent and currently account for more than 60% of total GHG emissions from rural areas. Considering the on-going construction of numerous new wastewater treatment systems in rural areas, enhancing wastewater treatment capacity, strengthening resource recovery, optimizing dietary patterns of the public, and promoting the use of clean energy are recommended to balance the trade-offs between environmental pollution abatement and climate change mitigation.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"28 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796456","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 : 2025-12-18DOI: 10.1038/s41545-025-00532-9
Weiyi Zhang, Weijie Chen, Jiwei Wang, Yang Xiao, Yunkai Li
Composite fouling, arising from microbial and inorganic interactions, compromises the reliability of reclaimed-water distribution systems. We evaluated permanent magnetic fields (PMF; 300–800 mT) and electromagnetic fields (EMF; AC and pulsed DC) in a full-scale reclaimed-water distribution loop. Under the tested conditions, EMF—particularly pulsed DC—reduced total foulant mass by up to 51.7% ( p < 0.05) and was associated with lower biofilm biomass, mineral scaling, and particulate accumulation. 16S rRNA profiling indicated co-occurring declines in community diversity and simpler co-occurrence networks under EMF. X-ray diffraction indicated a relative enrichment of aragonite (vs. calcite) and decreases in dolomite and quartz within deposits across all magnetic-field treatments. Field-induced oxidative conditions and interfacial changes were consistent with processes that may hinder microbial attachment and crystal nucleation. Collectively, the results suggest that EMF can outperform PMF for composite-fouling control in reclaimed-water systems and motivate optimization of field strength and waveform and validation across additional water qualities and hydraulics.
{"title":"Magnetic field mitigation of composite fouling through microbial and interfacial disruption","authors":"Weiyi Zhang, Weijie Chen, Jiwei Wang, Yang Xiao, Yunkai Li","doi":"10.1038/s41545-025-00532-9","DOIUrl":"https://doi.org/10.1038/s41545-025-00532-9","url":null,"abstract":"Composite fouling, arising from microbial and inorganic interactions, compromises the reliability of reclaimed-water distribution systems. We evaluated permanent magnetic fields (PMF; 300–800 mT) and electromagnetic fields (EMF; AC and pulsed DC) in a full-scale reclaimed-water distribution loop. Under the tested conditions, EMF—particularly pulsed DC—reduced total foulant mass by up to 51.7% ( <jats:italic>p</jats:italic> < 0.05) and was associated with lower biofilm biomass, mineral scaling, and particulate accumulation. 16S rRNA profiling indicated co-occurring declines in community diversity and simpler co-occurrence networks under EMF. X-ray diffraction indicated a relative enrichment of aragonite (vs. calcite) and decreases in dolomite and quartz within deposits across all magnetic-field treatments. Field-induced oxidative conditions and interfacial changes were consistent with processes that may hinder microbial attachment and crystal nucleation. Collectively, the results suggest that EMF can outperform PMF for composite-fouling control in reclaimed-water systems and motivate optimization of field strength and waveform and validation across additional water qualities and hydraulics.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"23 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145770815","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 : 2025-12-10DOI: 10.1038/s41545-025-00534-7
V. Campos, C. A. Tanaka, D. G. Marques, N. P. A. Granado
Freshwater scarcity demands materials and systems that enable efficient, low-cost, and environmentally responsible water generation. Here we report a hydrocell technology based on a recycled-polymer superabsorbent, poly(acrylamide-co-potassium acrylate) (PANSAP), synthesized through solvent-free alkaline hydrolysis of post-consumer polyacrylonitrile textiles. The process achieves an Environmental Factor (E-factor) of 0.056 when ammonia released during hydrolysis is recovered as ammonium phosphate, comparable to benchmark polymer-recycling efficiencies. The resulting cross-linked polymer exhibits a swelling capacity exceeding 200 g H₂O g⁻¹ and a stable atmospheric uptake of 0.43 ± 0.09 g g⁻¹ over 79 days at 69–90% relative humidity. Integrated into modular hydrocell plates, PANSAP delivers hybrid solar–electric desorption yields of 4–6 L day⁻¹ from 25 units at 0.90–1.25 kWh L⁻¹, approaching the thermodynamic minimum for water evaporation. Long-term cycling tests indicate negligible nitrogen loss (0.09 mg L⁻¹ NH₃ yr⁻¹), corresponding to an estimated service life beyond 2,500 cycles. Derived from recycled textiles, partially biodegradable, and agronomically beneficial through potassium release, PANSAP establishes a scalable, circular-materials framework for atmospheric water harvesting. The results position polymer-based hydrocells as a durable and sustainable platform for decentralized freshwater generation in arid and semi-arid regions.
{"title":"Scalable hydrocell technology based on recycled polymers for atmospheric water harvesting","authors":"V. Campos, C. A. Tanaka, D. G. Marques, N. P. A. Granado","doi":"10.1038/s41545-025-00534-7","DOIUrl":"https://doi.org/10.1038/s41545-025-00534-7","url":null,"abstract":"Freshwater scarcity demands materials and systems that enable efficient, low-cost, and environmentally responsible water generation. Here we report a hydrocell technology based on a recycled-polymer superabsorbent, poly(acrylamide-co-potassium acrylate) (PANSAP), synthesized through solvent-free alkaline hydrolysis of post-consumer polyacrylonitrile textiles. The process achieves an Environmental Factor (E-factor) of 0.056 when ammonia released during hydrolysis is recovered as ammonium phosphate, comparable to benchmark polymer-recycling efficiencies. The resulting cross-linked polymer exhibits a swelling capacity exceeding 200 g H₂O g⁻¹ and a stable atmospheric uptake of 0.43 ± 0.09 g g⁻¹ over 79 days at 69–90% relative humidity. Integrated into modular hydrocell plates, PANSAP delivers hybrid solar–electric desorption yields of 4–6 L day⁻¹ from 25 units at 0.90–1.25 kWh L⁻¹, approaching the thermodynamic minimum for water evaporation. Long-term cycling tests indicate negligible nitrogen loss (0.09 mg L⁻¹ NH₃ yr⁻¹), corresponding to an estimated service life beyond 2,500 cycles. Derived from recycled textiles, partially biodegradable, and agronomically beneficial through potassium release, PANSAP establishes a scalable, circular-materials framework for atmospheric water harvesting. The results position polymer-based hydrocells as a durable and sustainable platform for decentralized freshwater generation in arid and semi-arid regions.","PeriodicalId":19375,"journal":{"name":"npj Clean Water","volume":"6 1","pages":""},"PeriodicalIF":11.4,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145711576","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}
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}
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：