Abstract Researchers and engineers employ machine learning (ML) tools to detect pipe bursts and prevent significant non-revenue water losses in water distribution systems (WDS). Nonetheless, many approaches developed so far consider a fixed number of sensors, which requires the ML model redevelopment and collection of sufficient data with the new sensor configuration for training. To overcome these issues, this study presents a novel approach based on Long Short-Term Memory neural networks (NNs) that leverages transfer learning to manage a varying number of sensors and retain good detection performance with limited training data. The proposed detection model first learns to reproduce the normal behavior of the system on a dataset obtained in burst-free conditions. The training process involves predicting flow and pressure one-time step ahead using historical data and time-related features as inputs. During testing, a post-prediction step flags potential bursts based on the comparison between the observations and model predictions using a time-varied error threshold. When adding new sensors, we implement transfer learning by replicating the weights of existing channels and then fine-tune the augmented NN. We evaluate the robustness of the methodology on simulated fire hydrant bursts and real-bursts in 10 district metered areas (DMAs) of the UK. For real bursts, we perform a sensitivity analysis to understand the impact of data resolution and error threshold on burst detection performance. The results obtained demonstrate that this ML-based methodology can achieve Precision of up to 98.1% in real-life settings and can identify bursts, even in data scarce conditions.
{"title":"Leveraging Transfer Learning in LSTM Neural Networks for Data-Efficient Burst Detection in Water Distribution Systems","authors":"Konstantinos Glynis, Zoran Kapelan, Martijn Bakker, Riccardo Taormina","doi":"10.1007/s11269-023-03637-3","DOIUrl":"https://doi.org/10.1007/s11269-023-03637-3","url":null,"abstract":"Abstract Researchers and engineers employ machine learning (ML) tools to detect pipe bursts and prevent significant non-revenue water losses in water distribution systems (WDS). Nonetheless, many approaches developed so far consider a fixed number of sensors, which requires the ML model redevelopment and collection of sufficient data with the new sensor configuration for training. To overcome these issues, this study presents a novel approach based on Long Short-Term Memory neural networks (NNs) that leverages transfer learning to manage a varying number of sensors and retain good detection performance with limited training data. The proposed detection model first learns to reproduce the normal behavior of the system on a dataset obtained in burst-free conditions. The training process involves predicting flow and pressure one-time step ahead using historical data and time-related features as inputs. During testing, a post-prediction step flags potential bursts based on the comparison between the observations and model predictions using a time-varied error threshold. When adding new sensors, we implement transfer learning by replicating the weights of existing channels and then fine-tune the augmented NN. We evaluate the robustness of the methodology on simulated fire hydrant bursts and real-bursts in 10 district metered areas (DMAs) of the UK. For real bursts, we perform a sensitivity analysis to understand the impact of data resolution and error threshold on burst detection performance. The results obtained demonstrate that this ML-based methodology can achieve Precision of up to 98.1% in real-life settings and can identify bursts, even in data scarce conditions.","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136012558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-12DOI: 10.1007/s11269-023-03634-6
Chih-Hsien Lin, Wei-Hsiang Chen
{"title":"A Technology-Organization-Environment (TOE) Framework Based on Scientometry for Understanding The Risk Factors in Sustainable Water Resources Management","authors":"Chih-Hsien Lin, Wei-Hsiang Chen","doi":"10.1007/s11269-023-03634-6","DOIUrl":"https://doi.org/10.1007/s11269-023-03634-6","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136012830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-09DOI: 10.1007/s11269-023-03639-1
Hui Yue, Xiangyu Yu, Ying Liu, Xu Wang
{"title":"The Construction and Migration of a Multi-source Integrated Drought Index Based on Different Machine Learning","authors":"Hui Yue, Xiangyu Yu, Ying Liu, Xu Wang","doi":"10.1007/s11269-023-03639-1","DOIUrl":"https://doi.org/10.1007/s11269-023-03639-1","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135095511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Stable Improved Dynamic Programming Method: An Efficient and Accurate Method for Optimization of Reservoir Flood Control Operation","authors":"Fuxin Chai, Feng Peng, Hongping Zhang, Wenbin Zang","doi":"10.1007/s11269-023-03622-w","DOIUrl":"https://doi.org/10.1007/s11269-023-03622-w","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135254322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-06DOI: 10.1007/s11269-023-03623-9
Shiva Kandpal, Swaroop Nandan Bora
{"title":"Analytical Solution for Linearized Saint-Venant Equations with a Uniformly Distributed Lateral Inflow in a Finite Rectangular Channel","authors":"Shiva Kandpal, Swaroop Nandan Bora","doi":"10.1007/s11269-023-03623-9","DOIUrl":"https://doi.org/10.1007/s11269-023-03623-9","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135351324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-06DOI: 10.1007/s11269-023-03600-2
Chenchen Zhao, Chengshuai Liu, Wenzhong Li, Yehai Tang, Fan Yang, Yingying Xu, Liyu Quan, Caihong Hu
{"title":"Simulation of Urban Flood Process Based on a Hybrid LSTM-SWMM Model","authors":"Chenchen Zhao, Chengshuai Liu, Wenzhong Li, Yehai Tang, Fan Yang, Yingying Xu, Liyu Quan, Caihong Hu","doi":"10.1007/s11269-023-03600-2","DOIUrl":"https://doi.org/10.1007/s11269-023-03600-2","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135350303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-06DOI: 10.1007/s11269-023-03598-7
Wei Li, Xiaosheng Wang, Ran Li
{"title":"Water-Saving Revenue Guarantee Optimization in Water Saving Management Contract Based on Simulation Method","authors":"Wei Li, Xiaosheng Wang, Ran Li","doi":"10.1007/s11269-023-03598-7","DOIUrl":"https://doi.org/10.1007/s11269-023-03598-7","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135350540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-05DOI: 10.1007/s11269-023-03636-4
Alessio Nicosia, Costanza Di Stefano, Maria Angela Serio, Vito Ferro
Abstract The deduction of the weir flow stage-discharge relationship is a hydraulic problem generally solved by energy considerations and using the discharge coefficient to correct the gap between theoretical results and experimental measurements. In this context, the dimensional analysis represents an alternative to find simple and reliable equations to obtain the rating curve. In this study, the outflow process of vegetated weirs is investigated applying the Π-Theorem of dimensional analysis and the incomplete self-similarity theory. The aim of this paper is to propose a new theoretically-based stage-discharge relationship, and test its applicability by measurements recently published in the literature. The results showed that the errors in discharge estimate obtained by the proposed stage-discharge relationship are always less than or equal to ± 10% and less than or equal to ± 5% for 97–100% of cases. The main advantage of the proposed relationships is providing a single stage-discharge relationship, which has better performances than the equations reported in the literature and excludes the use of discharge coefficient.
{"title":"Dimensional Analysis and Stage-Discharge Relationships for Vegetated Weirs","authors":"Alessio Nicosia, Costanza Di Stefano, Maria Angela Serio, Vito Ferro","doi":"10.1007/s11269-023-03636-4","DOIUrl":"https://doi.org/10.1007/s11269-023-03636-4","url":null,"abstract":"Abstract The deduction of the weir flow stage-discharge relationship is a hydraulic problem generally solved by energy considerations and using the discharge coefficient to correct the gap between theoretical results and experimental measurements. In this context, the dimensional analysis represents an alternative to find simple and reliable equations to obtain the rating curve. In this study, the outflow process of vegetated weirs is investigated applying the Π-Theorem of dimensional analysis and the incomplete self-similarity theory. The aim of this paper is to propose a new theoretically-based stage-discharge relationship, and test its applicability by measurements recently published in the literature. The results showed that the errors in discharge estimate obtained by the proposed stage-discharge relationship are always less than or equal to ± 10% and less than or equal to ± 5% for 97–100% of cases. The main advantage of the proposed relationships is providing a single stage-discharge relationship, which has better performances than the equations reported in the literature and excludes the use of discharge coefficient.","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134976304","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-04DOI: 10.1007/s11269-023-03627-5
Zafarjon Sultonov, Hari K. Pant
{"title":"Potential Impacts of Climate Change on Water Management in the Aral Sea Basin","authors":"Zafarjon Sultonov, Hari K. Pant","doi":"10.1007/s11269-023-03627-5","DOIUrl":"https://doi.org/10.1007/s11269-023-03627-5","url":null,"abstract":"","PeriodicalId":23611,"journal":{"name":"Water Resources Management","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135591337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","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：