ML Water Usage Anomaly Detection via Property Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional water usage monitoring techniques lack context and effectiveness in detecting anomalous water usage at properties, failing to provide insights into conservative, average, or excessive usage, and often do not verify actual usage against community standards, potentially leading to unnoticed leaks or inefficiencies.
Innovation Solution
A computer-implemented method and system using machine learning (ML) that compares actual water usage data with estimated usage based on property attributes, generating notifications and actionable insights to property owners, utilizing IoT sensors, ML models trained on historical data, and providing context through water usage dashboards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional water usage monitoring is provided to customers, then customers receive water usage information, but the information lacks context and is of little use without comparison to community standards
Solution Approach 1:
The patent introduces an intermediary system comprising a server and database that mediates between water meters and customer devices. This intermediary collects raw water usage data, compares it against community benchmarks and historical data, and transforms it into contextualized insights. The server acts as the mediator that adds the missing context by performing comparative analysis against aggregated community data, thereby resolving the contradiction between providing useful information and maintaining system simplicity.
Solution Approach 2:
The system implements feedback mechanisms where water usage information is continuously collected, analyzed against community standards, and returned to customers in the form of contextualized reports. The feedback loop includes comparing individual property usage against community averages, identifying anomalies, and providing actionable insights. This feedback structure transforms raw data into meaningful information that helps customers understand their consumption patterns relative to community standards without requiring complex customer-side analysis.
2Measurement precision
If additional monitoring equipment is deployed to verify actual water usage, then measurement accuracy improves, but system complexity and cost increase
Solution Approach 1:
The patent employs universal water meters that serve multiple functions: they measure water usage, communicate data wirelessly, and store consumption information. These multi-functional meters eliminate the need for separate monitoring equipment by integrating measurement, communication, and data storage capabilities into a single device. This universality approach maintains measurement precision while avoiding the complexity increase that would result from deploying additional specialized monitoring equipment.
Solution Approach 2:
The system replaces complex mechanical monitoring infrastructure with electronic and wireless communication technologies. Instead of using mechanical data collection methods requiring manual intervention or complex mechanical sensors, the patent utilizes electronic water meters with integrated communication modules that transmit data wirelessly to the server. This substitution of mechanical systems with electronic alternatives maintains measurement accuracy while significantly reducing system complexity and operational overhead.
3Productivity
If water usage data is collected and analyzed without contextual comparison, then data collection is simple, but customers cannot identify anomalous usage or potential leaks
Solution Approach 1:
The system performs preliminary actions by pre-collecting and storing community benchmark data, historical usage patterns, and property attribute information in the database before analysis is needed. This preliminary preparation of reference data allows the server to quickly compare individual property usage against established benchmarks without performing complex real-time calculations. By having community standards and historical data pre-aggregated and stored, the system maintains high data processing efficiency while enabling reliable anomaly detection through straightforward comparative analysis.
Data Source
AI summary
System and methods for detecting anomalous water usage at a property having a water supply system using machine learning (ML) are provided. An ML model may generate an estimated water usage of the property associated with the property data set, where the property data set may include attributes of the property influencing the water usage of the property. The ML model may be configured to learn relationships between a historical water usage data set and a historical property data set. The systems and methods may compare the water usage of the property and the estimated water usage of the property. If the comparison indicates the water usage exceeds a threshold, a notification may be generated and provided to a user device.