PIPENET A Wireless Sensor Network for Pipeline Monitoring

PIPENET A Wireless Sensor Network for Pipeline Monitoring Ivan Stoianov Timur Tokmouline Imperial College, London CSAIL Lama Nachman Intel Research Sam Madden, MIT -Anandatirtha

Motivation US water companies are under increasing pressure to improve the management of their ageing assets and optimize operational and capital expenditure. Transmission and distribution projects represent the largest component ( 184 billion). The threat of contaminant intrusion due to leaking pipes or malicious human action will further increase the projected expenditure. Need to identify critical areas. Failures such as water leaks lead to dire consequences including loss of life, severe interruptions in service, degraded fire fighting ability, damage to infrastructure, and multi-million dollar repair bills.

PIPENET offers : Detect, localize, quantify bursts and leaks, detect blockages and detect malfunctioning of equipment such as control valves. Monitor quality of water transmission. Monitor water level in sewer collectors.

Known issues Pipelines are subject to complex, highly non- linear temporal and spatial processes that make it difficult to differentiate between faults and stochastic system behaviors. This makes detecting failures a challenging task.

PIPENET Working and Design Integrate and correlate data from several sources. Type of data monitored are acoustic/vibration signals, velocity (flow) signals, and pressure transient signals. Acoustic/vibration signals are used for detecting small leaks. Analysis of pressure transients and velocity (flow) enables prompt detection and localization of larger leaks and malfunctioning equipment such as air valves.

PipeNet Deployment.

Sensor Architecture Intel Mote Sensor Node Consists of an ARM7 core, 64kB RAM, 512kB Flash and Bluetooth radio. sensor board designed to interface the Intel Mote to various analog sensors used in PipeNet. The sensor board supports up to 8 analog channels.

Other design parameters Bluetooth scatternet formation and a tree routing algorithms to enable self configuring, self healing networks. Lightweight reliable transport protocol to support fragmentation and assembly of large data packets, as PipeNet motes frequently need to transfer messages of up to 100 kB.

Data Collection

Data Analysis Data from sensors is relayed via a GPRS modem to a backend server. Data from each sensor is loaded to a database. Data from sensors is analyzed. Possible integration with other applications like Google Maps, Web server etc are also possible.

Design Challenges Resilience to harsh environmental conditions. Separating data collection from communication. Time synchronization.