Outdoor Leak Detection Using Infrared Imagery and Meter Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Outdoor leaks in water supply networks are difficult to detect and locate due to the limited effectiveness of traditional monitoring methods, which can result in significant water loss and revenue loss, as well as property damage.
Innovation Solution
A system and method that combines environmental imagery, such as infrared imaging, with utility supply network data and property lot polygons using machine learning algorithms to identify and quantify leak locations, and includes automatic shutoff valve control to mitigate leaks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional monitoring methods are used to detect leaks in water supply networks, then the system can monitor utility supply amounts, but the ability to identify specific leak locations is limited
Solution Approach 1:
The patent combines multiple data sources including environmental imagery, utility supply network data, property lot polygons, and machine learning algorithms into an integrated leak detection system. This merging of previously separate monitoring approaches enables precise leak location identification while leveraging existing infrastructure rather than adding complex standalone systems.
Solution Approach 2:
The patent introduces an intermediary analysis system that processes environmental imagery and correlates it with utility network data. This intermediary layer translates raw imagery and meter data into actionable leak location information, bridging the gap between traditional monitoring and precise leak detection without requiring direct modification of the existing utility infrastructure.
2Loss of information
If utility monitoring is used to detect leaks, then the amount of leakage can be detected, but the location of the leak cannot be identified
Solution Approach 1:
The patent adds a spatial dimension to traditional utility monitoring by incorporating environmental imagery and geographic data. Instead of only monitoring flow amounts (one dimension), the system overlays visual spatial information from imagery with network data, enabling identification of leak locations in addition to detecting water loss amounts.
Solution Approach 2:
The system implements feedback by continuously comparing environmental imagery data with utility supply network data and meter readings. When discrepancies are detected between expected and actual water usage patterns correlated with environmental changes, the system provides feedback that identifies potential leak locations, enabling targeted investigation and remediation.
3Productivity
If manual leak detection methods are used in outdoor water supply networks, then leaks can be identified, but the process is costly and time consuming
Solution Approach 1:
The patent implements a self-service leak detection system that automatically monitors environmental imagery and utility data without requiring continuous manual inspection. The machine learning algorithms autonomously analyze the data, identify anomalies, and pinpoint leak locations, enabling the system to detect and report leaks without human intervention while significantly reducing detection time and costs.
Solution Approach 2:
The system performs preliminary analysis of environmental imagery and utility data continuously to identify potential leak conditions before significant water loss occurs. By proactively monitoring and analyzing data in real-time, the system detects leaks early in their development, enabling rapid response and remediation before the problem escalates and causes extensive damage.
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
This approach enables precise identification and notification of leak locations, reducing water loss and property damage by correlating environmental imagery changes with meter data and flow measurements, and automatically controlling supply flow to address detected leaks.
Implementation Method 1
infrared environmental imagery detects heat energy which can indicate wet areas on land surfaces
Data Source
AI summary
A computer implemented method implemented on a computer system including non-transient memory storing instructions for identifying leak locations based on environmental includes tracking a utility supply usage environment in environmental imagery captured over a period of time, identifying changes in the environmental imagery indicative of a leak location, applying a property profile to the utility supply usage environment to generate leak location property data, identifying one or more flow measurement devices based on the leak location and leak location property data, and generating a leak notification based on the leak location in the environmental imagery, the property profile data and meter data from the identified one or more flow measurement devices