A Thermal Imaging Based Wireless Sensor Network for Automatic Water Leakage Detection in Distribution Pipes

Water leakage is a significant problem in both developing and developed countries causing water loss in water-distribution systems. Leakage causes economic loss in the form of wastage of water, damage to pipe networks and foundations of roads and buildings, and also poses risk to public health due to water contamination. The lost or unaccounted amount of water is typically 20-30 percent of production. Some older systems may lose even up to 50 percent. The water pipe networks in houses as well as public places are generally concealed and hence detecting water leakage in the initial stages before an upcoming damage is difficult. The existing technologies for detecting leakage have various limitations such as efficiency being dependent on size, material and depth of pipes, need for manual intervention, dependency on weather and surface conditions and effect of water pressure. In this paper, we propose an automated water leakage detection system using a wireless sensor network created along the distribution pipes based on thermal (IR) imaging. Thermal imaging has the capability to work in low lighting or dark conditions and helps in capturing the contrast between hot and cold areas created due to water leakage. A network of low cost, low power thermal imaging sensors each having its own processing and Radiofrequency (RF) Transreceiver units and operating independent of pipe, weather or surface conditions is proposed in this paper. A central database is updated on a real-time basis, enabling very early leakage detection and initiating subsequent action to address the problem. An example system evaluation is performed and results highlighting the power and cost impact of the sensor network system are presented.