Design a sensor architecture combining acoustic loggers, pressure transducers and smart meters for continuous network monitoring.
Digital Transformation & Artificial Intelligence
AI-Based Leak Detection and Smart Network Management for Water Utilities
Use AI-based leak detection and smart network monitoring to cut non-revenue water and prioritise pipe repairs before failures happen.
🗓️Duration5 training days
📚Content4 modules · 8 sessions
🎓On completionAccredited attendance certificate
About the programme
Course Overview
Water utilities routinely lose a significant share of treated water to leaks before it ever reaches a paying customer, and traditional leak detection through acoustic surveys and customer reports finds problems only after water has already been lost or a pipe has burst. AI-based leak detection combines acoustic sensors, pressure transient monitoring, smart meter data and machine learning models to flag likely leaks earlier and prioritise which pipe sections to inspect first, turning a reactive maintenance function into a managed, data-driven programme. This course teaches water utility engineers and asset managers to design a sensor and data architecture for continuous network monitoring, build or evaluate machine learning models that distinguish genuine leak signatures from normal network noise, and integrate model output into a maintenance prioritisation and pressure management programme. Sessions cover sensor placement strategy, acoustic and pressure-based detection methods, model validation against confirmed leak locations, and the business case for a smart network programme against a utility's non-revenue water reduction targets. Participants leave with a network monitoring architecture, a leak prioritisation model design, and a business case they can present to utility leadership and regulators.
Expected Learning Outcomes
Distinguish genuine leak signatures from normal network noise using acoustic and pressure data analysis.
Evaluate or build machine learning models that prioritise pipe sections most likely to be leaking.
Validate leak detection model output against confirmed leak locations from field crews.
Integrate leak prioritisation output into maintenance scheduling and crew dispatch systems.
Design a pressure management strategy that reduces both leak volume and pipe failure frequency.
Build a business case linking a smart network programme to non-revenue water reduction targets and cost savings.
Who Should Attend
Water utility asset managers responsible for reducing non-revenue water and pipe failure rates.
Network engineers designing or upgrading acoustic and pressure monitoring infrastructure.
Data scientists building leak prioritisation models from utility sensor and meter data.
Operations managers responsible for scheduling maintenance crews and repair budgets.
Utility executives building a business case for smart network investment to present to a board or regulator.
Regulatory affairs staff reporting non-revenue water performance to a water sector regulator.
Course Modules
Select any module to see its sessions and points.
01Building the Sensor and Data Architecture
2 sessions · 8 points
Session 1Sensor Types and Network Instrumentation Strategy
- Compare acoustic loggers, pressure transducers and flow meters on their leak detection strengths and blind spots.
- Design sensor placement density that balances detection coverage against installation and maintenance cost.
- Integrate smart meter data as an additional signal for identifying unusual consumption patterns linked to leaks.
- Plan communication infrastructure, such as cellular or low-power wide-area networks, for remote sensor data collection.
Session 2Establishing a Reliable Data Pipeline for Network Monitoring
- Design a data pipeline that ingests acoustic, pressure and meter data into a common time-series platform.
- Set data quality checks that flag sensor faults, dropouts or miscalibration before they corrupt analysis.
- Establish a baseline of normal network noise and pressure behaviour against which anomalies are measured.
- Plan data retention and storage that supports both real-time alerting and long-term trend analysis.
02Detecting Leaks with Acoustic and Pressure Analytics
2 sessions · 8 points
Session 1Acoustic Signature Analysis for Leak Identification
- Identify the acoustic signatures that distinguish a genuine leak from traffic noise, valve operation or other interference.
- Apply correlation techniques between paired acoustic sensors to estimate leak location along a pipe segment.
- Tune detection sensitivity to balance false positives against missed leaks given crew investigation capacity.
- Validate acoustic detections against confirmed leaks found by field crews to calibrate model confidence.
Session 2Pressure Transient Monitoring and Machine Learning Models
- Detect pressure transients and gradual pressure decline patterns associated with developing leaks or pipe bursts.
- Train machine learning models on historical leak and non-leak data to score pipe sections by leak likelihood.
- Combine acoustic, pressure and consumption features into a single model to improve prioritisation accuracy.
- Evaluate model precision and recall against the utility's actual crew investigation capacity and cost per inspection.
03Prioritising Repairs and Managing Pressure
2 sessions · 8 points
Session 1Turning Model Output into a Repair Prioritisation Programme
- Rank flagged pipe sections by leak probability, estimated volume loss and consequence of failure if unrepaired.
- Integrate prioritised leak lists into maintenance scheduling and crew dispatch systems.
- Feed field investigation outcomes back into the model to improve future prioritisation accuracy.
- Set thresholds that trigger emergency response for high-confidence, high-consequence leak indications.
Session 2Pressure Management to Reduce Leakage and Failures
- Design pressure zones and pressure-reducing valve settings that lower background leakage without harming service levels.
- Model the relationship between reduced average pressure and both leak volume and pipe burst frequency.
- Coordinate pressure management changes with customer service level commitments and fire flow requirements.
- Monitor the effect of pressure management interventions on leak rates over subsequent months.
04Building the Business Case and Scaling the Programme
2 sessions · 8 points
Session 1Quantifying Non-Revenue Water Savings and Cost Avoidance
- Calculate the value of reduced non-revenue water in terms of treatment, energy and water resource cost avoided.
- Estimate cost avoidance from fewer emergency repairs and reduced infrastructure damage from pipe bursts.
- Compare the cost of sensor deployment and analytics against realised savings over a defined payback period.
- Present the business case in terms aligned with the utility's regulatory performance targets.
Session 2Scaling the Programme and Sustaining Performance
- Plan a phased rollout that expands sensor coverage from a pilot zone to the full network.
- Assign ownership for maintaining sensors, retraining models and updating baseline network behaviour.
- Report non-revenue water and leak detection performance to utility leadership and the sector regulator.
- Review the programme periodically to incorporate new sensor technology and improved detection models.
What the participant receives
4 course modules
A structured syllabus
8 training sessions
across 5 days
32 detailed points
Applied, detailed content
Accredited attendance certificate
On completing the programme
Complete your registration
We will contact you within one business day to confirm.
Ready to start?
Reserve your seat and start building the skill.