Acoustic Sensor for Unknown Water Detection in Sewer Pipes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for detecting unknown water in sewer systems, which includes rainwater and groundwater entering wastewater pipelines, are costly due to the need for flowmeter installations, and there is a need for a more economical solution.
Innovation Solution
A machine learning-based unknown-water detection apparatus that extracts acoustic feature patterns from sound data collected in sewer pipes, using a processor and memory to predict the presence of unknown water without the need for flowmeters, by learning patterns from acoustic data during non-rainfall and rainfall conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If flowmeters are installed to detect unknown water in sewer pipes, then measurement precision is improved, but device complexity and installation cost increase
Solution Approach 1:
The patent replaces mechanical flowmeters with an acoustic detection system that uses microphones and machine learning algorithms to detect unknown water. The acoustic sensor captures water flow sounds, and a neural network model analyzes these sounds to identify unknown water infiltration, eliminating the need for mechanical installation in the sewer pipe.
Solution Approach 2:
The patent creates a virtual model of normal water flow acoustic patterns through machine learning training, then compares actual acoustic patterns against this model to detect anomalies. This copying approach allows detection without physical intervention in the pipe system.
2Measurement precision
If flowmeters are installed to detect unknown water, then measurement capability is improved, but installation cost and labor cost increase
Solution Approach 1:
The patent replaces expensive mechanical flowmeters with acoustic sensors and software-based detection. The acoustic sensor unit is much cheaper to install, requiring only placement near the sewer access point rather than insertion into the pipe system.
Solution Approach 2:
The system uses ambient acoustic data and machine learning models that can be trained on-site or remotely updated, reducing the need for expensive professional installation and calibration services required by traditional flowmeters.
3Device complexity
If conventional acoustic detection methods are used without machine learning, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent creates a digital copy or model of normal water flow acoustic patterns through machine learning training data. This model serves as a reference for comparing actual acoustic patterns, enabling accurate detection of unknown water without complex manual analysis systems.
Solution Approach 2:
The system transforms raw acoustic data into extracted features (frequency spectrum, time-domain characteristics) that enhance the distinguishability of unknown water patterns. This parameter transformation enables simple comparison logic to achieve high detection accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the detection of unknown water at a lower cost than conventional methods, using acoustic data to identify temporal changes in water flow sounds, effectively predicting the presence of unknown water without requiring flowmeter installations.
Implementation Method 1
an acoustic sensor unit 12 that collects acoustic data including a water flowing sound in a sewer pipe
Implementation Method 2
an unknown-water prediction section 13 that predicts presence or absence of unknown water from an acoustic feature amount pattern of target acoustic data for prediction by utilizing a machine learning model
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Provided is a technology for realizing detection of unidentified water without using a flowmeter. One embodiment of the present invention pertains to an unidentified water detection device having: a feature quantity extraction unit which extracts, from acoustic data including running water sound, an acoustic feature quantity pattern that indicates temporal changes in an acoustic feature quantity; and an unidentified water prediction unit which, using a machine learning model learned from the acoustic feature quantity pattern extracted from data that contains acoustic data including running water sound during lack of rainfall and/or acoustic data of running water sound under a condition different from during lack of rainfall, predicts the presence/absence of unidentified water from the acoustic feature quantity pattern of acoustic data of a prediction object, wherein the acoustic feature quantity pattern indicates temporal changes in the acoustic feature quantity in terms of time period and day of the week.