Urban Pipe Network Digital Transformation: Data Silos and IoT Platforms
Yicai Technology Newsroom | Urban Pipe Network Industry Watch
Lead: From “Laying Pipes” to “Computing Systems”
Urban water supply pipe networks are shifting from incremental construction to intensive management of existing assets. Over the past two decades, the industry’s focus was on pipe diameter, length, coverage, and water supply capacity. Over the next decade, the focus will shift to leakage rate, pressure stability, compliance rate at water quality endpoints, average repair time for pipe bursts, and energy consumption per unit of water supplied.
The Implementation Plan for Leakage Control of Urban Water Supply Pipe Networks proposes that by 2025, the leakage rate of urban public water supply pipe networks nationwide should be controlled within 9%, and in some water-scarce cities within 7%. This target shifts pipe network governance from “emergency repair-oriented” to “system-oriented”.
Market signals are changing accordingly. Urban renewal, infrastructure lifeline safety projects, and resilient city pilot programs are directing funds from new pipe networks toward inspection, repair, monitoring, modeling, and O&M. A typical prefecture-level city water supply system often contains thousands of kilometers of pipe networks, dozens of pumping stations, hundreds of valves, and tens of thousands of user connection points.
Manual inspection alone can no longer meet fine-grained management requirements. The industry is beginning to accept a new paradigm that is “perceptible, modelable, predictable, and actionable”.
Market Trends: Reshaping Investment Logic
From an investment structure perspective, urban pipe network projects are presenting three overlapping categories:
- Renewal and Retrofitting: Focusing on corrosion, joint leakage, and burst risks of gray cast iron pipes, steel pipes, concrete pipes, and other materials.
- Strengthening Monitoring and Sensing Systems: Including district metered area (DMA) metering, pressure monitoring, flow monitoring, and online water quality monitoring.
- Digital Platforms and Model Services: Covering GIS, SCADA, hydraulic models, leakage analysis, and O&M work order systems.
Industry estimates show that over the next five years, investment in urban water supply and drainage pipe network renewal, retrofitting, and digitalization is expected to maintain a relatively high growth rate, with the share of asset investment related to leakage control rising continuously.
Assessment Methods Are Changing
In the past, some cities mainly looked at water supply capacity and water quality compliance rate. In the future, they are more likely to look at process indicators: whether pipe network pressure remains stable within a reasonable range, whether nighttime minimum flow is abnormal, whether pipe burst warnings are issued in advance, whether water outage repairs are rapid, and whether secondary water supply quality is controllable.
Case Study: A prefecture-level city in East China built a DMA and pressure monitoring system serving 600,000 people. By deploying approximately 2,200 monitoring points, the city reduced the leakage rate from 11.8% to 7.6%, saved about 2.6 million cubic meters of water annually, and shortened the average pipe burst handling time from 6.5 hours to 3.2 hours.
Leakage Analysis: Managing Leakage Structure
The core of leakage analysis has upgraded from “finding leak points” to “managing leakage structure”. The industry typically classifies pipe network leakage into four categories:
Background Leakage
Related to pressure, pipe age, material, joint type, and soil corrosion environment.
Joint Leakage
Mostly occurs at spigot-and-socket joints, flanges, valve wells, and aging joints.
Burst Leakage
Sudden events, but risk can be predicted through pressure fluctuation, water hammer, and corrosion models.
Metering Errors
Arise from large meters, water meters, pressure-flow sensors, and meter reading cycles.
Truly effective leakage control requires simultaneously addressing these four categories of issues, rather than simply relying on nighttime acoustic leak detection.
DMA Metering as the Basic Unit
In practice, a DMA typically covers 5,000 to 20,000 households, or a daily water supply volume of about 10,000 to 20,000 cubic meters. Nighttime minimum flow (MNF) is a commonly used indicator. When a DMA’s nighttime minimum flow exceeds the historical baseline by 15% to 20% for three consecutive days, the system can trigger an investigation work order.
After a northern city introduced acoustic correlators and AI recognition in aging pipe network areas, leak location error decreased from an average of 15 meters to within 5 meters, nighttime inspection efficiency improved by about 40%, and leakage control cost per DMA decreased by about 18%.
Pressure Monitoring: Zonal Pressure Optimization
Pressure monitoring is a key variable for coordinated leakage and water quality governance. Water supply pipe networks are not safer with higher pressure; excessively high pressure increases background leakage, joint leakage, and the probability of pipe bursts. Excessively low pressure affects secondary water supply, firefighting water use, and end-user service.
Key Insight: Nonlinear Relationship
Industry experience shows that background leakage volume and pressure often have a nonlinear relationship, with a pressure exponent between 1.15 and 1.5. This means that for every 0.1 MPa reduction in pressure, background leakage may decrease by 10% to 20%.
Technically, pressure monitoring points are usually deployed at pumping station outlets, trunk pipe nodes, secondary water supply inlets, zone boundaries, and high-risk pipe sections. Common pressure sensors have an accuracy of about 0.5% FS, with sampling frequencies of 1 to 5 minutes, and NB-IoT, LoRa, or 5G gateways upload data to platforms.
A water-scarce city, through peak pressure control, reduced trunk pipe pressure fluctuation amplitude from ±0.08 MPa to ±0.03 MPa, decreasing the pipe burst rate by about 30% within six months and reducing pumping station electricity consumption by about 6%.
Water Quality: Controllable Throughout the Process
Water quality monitoring is shifting from “compliance at the plant outlet” to “controllability throughout the process”. Risks arise not only from source water and treatment, but also from pipe network endpoints, secondary water supply, water outage and restoration, and biofilm.
- Conventional Online Indicators: Turbidity, residual chlorine, pH, temperature, conductivity, and pressure.
- Advanced Indicators: Ammonia nitrogen, dissolved oxygen, total chlorine, color, odor, and microbial risk proxy indicators.
- Critical Thresholds: Residual chlorine at pipe network endpoints usually needs to be maintained no lower than 0.05 mg/L, and turbidity is generally controlled within 1 NTU.
What truly matters is whether indicators are stable, whether abnormal fluctuations occur, and whether they are associated with water outages or pressure events. Water quality events often require minute-level early warning.
Scenario: A southern city deployed online water quality instruments at secondary water supply tanks. Combined hydraulic models identified a section of aging steel pipe with biofilm accumulation risk. Through a combined judgment of rising turbidity, declining residual chlorine, and abnormal conductivity, the system issued a cleaning recommendation 4 hours in advance, avoiding a water quality complaint affecting thousands of households.
Technology Directions: Ubiquitous Sensing & Dynamic Modeling
Ubiquitous Sensing
Expanding from traditional SCADA to low-power IoT sensors, edge gateways, acoustic leak detection, distributed fiber optic sensing, and mobile inspection devices.
Dynamic Modeling
Moving from static hydraulic models to digital twins. Emphasizing a closed data loop where GIS, SCADA, and anomaly events continuously calibrate roughness, demand patterns, and valve openings.
Autonomous Decision-Making
Edge computing reduces anomaly detection latency from minute-level to second-level, suitable for scenarios such as pipe bursts, water hammer, and sudden water quality changes.
After three months of online calibration, a city pipe network model reduced node pressure simulation error from 8% to within 3%, and the accuracy of pipe burst impact scope prediction improved to over 80%.
Policy Impact: Data Quality and Standard Unification
Policy impact is extending from “target-based assessment” to “data quality and standard unification”. Leakage rate control targets provide direction, but implementation requires unified calibers: how the leakage rate is calculated, how DMA boundaries are determined, how metering errors are excluded, how pressure data is accepted, and how water quality warnings are graded.
In the future, local standards may be refined to monitoring point deployment density, data upload frequency, anomaly response time limits, and model calibration cycles. For example, key DMAs may require pressure, flow, and water quality data to be uploaded at minute or hour intervals, pipe burst work orders confirmed within 30 minutes, and water outage information pushed to affected users within 1 hour.
Policy tools are becoming diversified. Water-saving contract management, EPC+O, urban lifeline safety projects, special funds for urban renewal, and water supply price reform are affecting project business models. The industry needs to avoid turning digitalization into new “data silos” by ensuring clear data ownership, verifiable quality, and controllable security.
Outlook: Resilience, Collaboration, and Autonomy
Looking forward over the next five years, urban pipe network governance will show three clear trends:
- From Leakage Control to Water System Resilience: Reduced leakage rate is only one outcome; what matters more is whether the system can maintain core services in the face of pipe bursts, drought, pollution, extreme rainfall, and power disturbances.
- From Single-Point Monitoring to Full-Domain Collaboration: Pressure, flow, water quality, energy consumption, GIS, and work orders will enter the same decision-making framework. Water supply pipe networks will no longer operate in isolation but will be coordinated with pumping stations, secondary water supply, firefighting, drainage, and energy systems.
- From Manual Experience to Model Autonomy: By 2030, more prefecture-level and above cities are expected to establish dynamic digital twin water supply systems; key DMAs will achieve joint pressure-flow-water quality early warning, and AI models will be able to automatically generate recommendations for leakage investigation, valve closure for pipe bursts, and user notifications.
