Leaksense-Sensormine Platform

• Enables daily, weekly or monthly view of alert types and numbers with field feedback cycle enabling utility analytic customisation.

• Integrates with utility geospatial, work order and business reporting systems for alert response management and auditing.

• Developed within an Azure cloud based SaaS platform, offering out-of-the-box integration for any industry partners.

• Works well with different levels of users from experienced operators to non-technical users, easy to implement and update.

Sensor Installation and Diagnostics

• Sensor installation (including placement) to optimise sensitivity, reduce artifacts and achieve reliable performance.

• Sensor sensitivity, filtering and, amongst other things, dynamic gain behaviour testing.

• Analysis to determine whether sensor field configurations have been disrupted and whether action is required.

• Fingerprinting of sensor operational histories for visualisation and checking of work order records.

• Work order system integrations and feedback loops for improvement through to auditing.

Event Classification

• Leak identification and characterisation using statistical and machine learning analytics.

• Analysis conducted in the audio and visual spectrum over time and frequency domains.

• Algorithms developed continuously over 5 years using field verification data feedback.

• Alert priorities determined using leak versus environmental noise discriminators and statistical and machine learning driven logic.

Pattern Recognition

• Early leak and pipe break detection using sensitive frequency power level change analysis.

• On-going tracking of leak and pipe break development and repair prioritisation.

• Identification of other regular or irregular patterns and discrimination from leaks.

• Time and space change and pattern recognition for single, double or greater numbers of sensing locations combined with characterisation of individual signals.

Denoising and Localisation

• Acoustic signal reinforcement using denoising methods to focus on leak sources and enable earlier detection.

• Use of data from multiple sensors as part of signal reinforcement.

• Focus on coherent acoustic energy to improve leak localisation accuracy.

• Use of coherent acoustic energy for leak characterisation and prioritisation.