Comparison of model-based techniques for pipe burst location in water distribution networks

Ferreira, Bruno; Antunes, André; Carriço, Nelson; Covas, Dídia

doi:10.1088/1755-1315/1136/1/012039

Cited by 3 publications

(3 citation statements)

References 31 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As such, when the uncertainty is significant, those methods can pinpoint to completely wrong locations. Ferreira et al [ 34 ] compared three distinct model-based techniques for this same case study and concluded that given high uncertainty levels (e.g., noise level of 3%), the unique inferred location is often quite distant from the true location (e.g., between 200 and 300 m). This unique location is not an ideal result since, if wrongly inferred, the water utility may be looking for a pipe burst where there is not one.…”

Section: Application: Results and Discussionmentioning

confidence: 99%

Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks

et al. 2023

Self Cite

View full text Add to dashboard Cite

The current paper presents a hyper parameterization optimization process for a convolutional neural network (CNN) applied to pipe burst locations in water distribution networks (WDN). The hyper parameterization process of the CNN includes the early stopping termination criteria, dataset size, dataset normalization, training set batch size, optimizer learning rate regularization, and model structure. The study was applied using a case study of a real WDN. Obtained results indicate that the ideal model parameters consist of a CNN with a convolutional 1D layer (using 32 filters, a kernel size of 3 and strides equal to 1) for a maximum of 5000 epochs using a total of 250 datasets (using data normalization between 0 and 1 and tolerance equal to max noise) and a batch size of 500 samples per epoch step, optimized with Adam using learning rate regularization. This model was evaluated for distinct measurement noise levels and pipe burst locations. Results indicate that the parameterized model can provide a pipe burst search area with more or less dispersion depending on both the proximity of pressure sensors to the burst or the noise measurement level.

show abstract

Section: Application: Results and Discussionmentioning

confidence: 99%

Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…The identification of such critical areas can be carried out by simulating a pipe burst event (i.e., using the hydraulic simulation model) in distinct network locations and by attempting to locate such simulated bursts in the network using a given advanced algorithm, for instance, inverse analysis, sensitivity-based methods, or classifier-based methods [29]. If the simulated event is correctly located, it means that a real event occurring in such a location can also be potentially located.…”

Section: Identification Of Critical Areas For Pipe Burst Locationmentioning

confidence: 99%

“…The pressure measurements generated by such a simulated event are then used by the advanced algorithm to locate such an event. The implemented advanced algorithm is based on a pattern recognition classifier (further details are presented in [29]), but distinct methods could be used (e.g., inverse analysis). The assessment of the correct burst location is given by the distance between the identified location and the correct location, measured along the network pipes (the shortest distance) or given by the Euclidean distance.…”

Section: Identification Of Critical Areas For Pipe Burst Locationmentioning

confidence: 99%

Computational Tools for Supporting the Operation and Management of Water Distribution Systems towards Digital Transformation

Carriço

Ferreira

Antunes

et al. 2023

Water

Self Cite

View full text Add to dashboard Cite

This paper presents a set of computational tools specially developed for supporting the operation and management of water distribution systems towards digital transformation of water services. These tools were developed in the scope of two R&D projects carried out in Portugal, DECIdE and WISDom, during 2018–2022. The DECIdE project focused on the development of tools for importing cadastral and operational data, as well as on the three operational tools for supporting the performance assessment: the first allows the calculation of different key performance indicators, both at a global and sectorial level, which is an annual requirement of the water regulator, and the other two allow the calculation of the water and the energy balances and a set of complementary indices. The WISDom project aimed at the implementation of applications that directly address specific water utility needs, namely, the flow rate data processing, the optimal location of pressure sensors, the identification of critical areas in the distribution network for pipe burst location, and the prioritization of pipes for rehabilitation. Implemented tools are useful to support water utilities in the daily operation and management of their systems, being a step forward towards digital transformation of the water sector.

show abstract

Comparative Analysis of Leak Detection Methods Using Hydraulic Modelling and Sensitivity Analysis in Rural and Urban–Rural Areas

Bartkowska,

Wysocki,

Zajkowski

et al. 2024

Sustainability

View full text Add to dashboard Cite

Water scarcity is a significant global challenge, exacerbated by leakages in water distribution networks. This paper addresses the challenge of detecting leakages in rural and urban–rural water supply systems through hydraulic modelling and a sensitivity analysis. Two distinct real-world network models were studied to assess real and simulated leakage scenarios varying in location and magnitude. A distinct leakage detection approach utilizing outflow measurements from hydrants was tested. Additionally, the effectiveness of various statistical measures—such as correlation, angular closeness, Euclidean distance, Manhattan distance, Chebyshev distance, cosine similarity, and Spearman correlation—were evaluated to determine their efficacy in leakage detection. Different methods for identifying leak candidates were explored and compared, either by selecting a single leak candidate based on similarity measures or by identifying a group of candidates to mark leak hotspots. Density-based spatial clustering of applications with noise was used to assess the number of potential leak candidate groups. The study’s findings contribute to the optimization of leak detection strategies in water supply networks, particularly in rural settings, where detection is challenging due to scarce measurement datasets, budget restrictions, and operational constraints.

show abstract

Comparison of model-based techniques for pipe burst location in water distribution networks

Cited by 3 publications

References 31 publications

Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks

Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks

Computational Tools for Supporting the Operation and Management of Water Distribution Systems towards Digital Transformation

Comparative Analysis of Leak Detection Methods Using Hydraulic Modelling and Sensitivity Analysis in Rural and Urban–Rural Areas

Contact Info

Product

Resources

About