article · Sensors
Water leakages in building pipelines contribute substantially to water waste, particularly in water-stressed areas facing strain from ageing infrastructure. Addressing this challenge requires efficient, automated detection systems. An edge computing approach enables local, real-time water management by running compact machine learning models directly on connected sensors, reducing reliance on centralised processing. Using an existing acoustic dataset recorded from polyvinyl chloride pipelines, acoustic signals were converted into scalograms to train five convolutional neural network architectures through transfer learning. Among the evaluated models, EfficientNet delivered the highest performance, achieving an accuracy of 97.45 percent and an F1 score of 97.63 percent. When compressed using quantisation for deployment on an Arduino Nano 33 BLE edge microcontroller, the model operated with an inference time of 1932 milliseconds, requiring 255.3 kilobytes of peak RAM and 48.7 kilobytes of flash storage.
Unnoticed pipe leakages drive severe water losses in cities and water-scarce regions. Implementing low-power machine learning models directly onto inexpensive hardware provides real-time monitoring without continuous network connectivity or high computing costs. This decentralised monitoring method offers a practical route to conserving water resources and maintaining ageing pipe infrastructure in residential and commercial buildings.
The research demonstrates an applied prototype tested on physical hardware, specifically an Arduino Nano 33 BLE device using acoustic pipe data. This capability could be integrated into building management systems or commercialised by smart meter and Internet of Things hardware vendors for real-time monitoring of polyvinyl chloride plumbing. Because the solution operates on low-cost microcontrollers with minimal memory requirements, it is technically close to field testing in operational building environments.
AI-generated from the published abstract. Always read the original work before citing.
The escalating global water usage and the increasing strain on major cities due to water shortages highlights the critical need for efficient water management practices. In water-stressed regions worldwide, significant water wastage is primarily attributed to leakages, inefficient use, and aging infrastructure. Undetected water leakages in buildings' pipelines contribute to the water waste problem. To address this issue, an effective water leak detection method is required. In this paper, we explore the application of edge computing in smart buildings to enhance water management. By integrating sensors and embedded Machine Learning models, known as TinyML, smart water management systems can collect real-time data, analyze it, and make accurate decisions for efficient water utilization. The transition to TinyML enables faster and more cost-effective local decision-making, reducing the dependence on centralized entities. In this work, we propose a solution that can be adapted for effective leakage detection in real-world scenarios with minimum human intervention using TinyML. We follow an approach that is similar to a typical machine learning lifecycle in production, spanning stages including data collection, training, hyperparameter tuning, offline evaluation and model optimization for on-device resource efficiency before deployment. In this work, we considered an existing water leakage acoustic dataset for polyvinyl chloride pipelines. To prepare the acoustic data for analysis, we performed preprocessing to transform it into scalograms. We devised a water leak detection method by applying transfer learning to five distinct Convolutional Neural Network (CNN) variants, which are namely EfficientNet, ResNet, AlexNet, MobileNet V1, and MobileNet V2. The CNN models were found to be able to detect leakages where a maximum testing accuracy, recall, precision, and F1 score of 97.45%, 98.57%, 96.70%, and 97.63%, respectively, were observed using the EfficientNet model. To enable seamless deployment on the Arduino Nano 33 BLE edge device, the EfficientNet model is compressed using quantization resulting in a low inference time of 1932 ms, a peak RAM usage of 255.3 kilobytes, and a flash usage requirement of merely 48.7 kilobytes.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/s23229210
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
