Water Flow Detection Device Based on Sound Data Analysis and Machine Learning to Detect Water Leakage

TL;DR

This work presents a low-cost, non-invasive water-leak detection system that attaches to the exterior of building pipes and uses a mechanical sound amplifier to boost leakage-related acoustics captured by a microphone. After converting sounds to digital signals, the study extracts multiple features (STFT, DWT, MFCCs, FBANK) and finds FBANK features most effective for discrimination. A deep Gaussian RBF (G-RBF) classifier with SoftML achieves the highest accuracy (up to 97.7% F1) on nine datasets, with strong generalization demonstrated through fine-tuning across three real locations. The approach enables reliable detection of leaks exceeding 100 mL/min and offers a scalable, non-invasive solution for monitoring water infrastructure in diverse settings.

Abstract

In this paper, we introduce a novel mechanism that uses machine learning techniques to detect water leaks in pipes. The proposed simple and low-cost mechanism is designed that can be easily installed on building pipes with various sizes. The system works based on gathering and amplifying water flow signals using a mechanical sound amplifier. Then sounds are recorded and converted to digital signals in order to be analyzed. After feature extraction and selection, deep neural networks are used to discriminate between with and without leak pipes. The experimental results show that this device can detect at least 100 milliliters per minute (mL/min) of water flow in a pipe so that it can be used as a core of a water leakage detection system.

Figures (9)

• Figure 1: An overview of the proposed water leakage system detection.
• Figure 2: A sample of an acoustic sensor used in the proposed water flow detection system.
• Figure 3: Proposed system (a) CAD model; (b) Prototype.
• Figure 4: Processing block diagram to detect water flow.
• Figure 5: Visualization of different flows in two dimensions using the t-SNE algorithm (a) comparing the flow of 50 mL/min with zero flow; (b) comparing the flow of 100 mL/min with zero flow; (c) comparing the flow of 250 mL/min with zero flow.