A Regina-based tech startup says it's the first company to use artificial intelligence (AI) to detect leaks at oil wells and pump jacks. In the past, oil and gas companies have used staff to drive to oil wells to inspect them for any issues, such as leaks. One solution is using remote cameras to monitor oil wells, but it results in hundreds or thousands of photos being taken that have to be inspected by employees. Founded in 2018, Wave9 takes the arduous task of inspecting those photos and hands it off to AI. Cameras and sensors placed on pump jacks are processed by the software. The user can then be alerted to issues through apps that run on tablets and smartphones.

Swarms of floating robots could help map underground pipe networks and detect leaks and blockages in plumbing. Peter Baltus at Eindhoven University of Technology in the Netherlands and his colleagues have developed golf ball-sized sensors that can collect information as they float through pipes. Each robot contains a microprocessor, sensor, memory boards and a battery. They can be programmed to detect sound, temperature, pressure, acceleration, rotation and magnetic fields. To save power, a sensor can be activated by a sudden change in conditions, such as hissing sounds associated with water escaping, or increased rotation, which could be a sign of turbulent water flow.

The timely detection of leaks in water distribution systems is critical to the sustainable provision of clean water to consumers. Increasingly, water companies are deploying remote sensors to measure water flow in real-time in order to detect such leaks. However, in practice, for typical District Metering Zones (DMZ), financial constraints limit the number of deployable real time flow sensors/meters to one or two, thus constraining leak detection to be based on the aggregated flow being monitored at these point. Such aggregated flow data typically exhibits input signal dependence whereby both noise and leaks are dependent on the flow being measured. This limited monitoring and input signal dependance make conventional approaches based on simple thresholds unreliable for real time leak detection. To address this, we propose a Gaussian process (GP) model with an additive diagonal noise covariance that is able to handle the input dependant noise observed in this setting. A parameterised mean step change function is used to detect leaks and to estimate their size. Using prior water distribution systems (WDS) knowledge we dynamically bound and discretize the detection parameters of the step change mean function, reducing and pruning the parameter search space considerably. We evaluate the proposed noise scaled GP (NSGP) against both the latest researchwork on GP based fault detection methods and the current state of the art and applied leak detection approaches in water distribution systems. We show that our proposed method outperforms other approaches, on real water network data with synthetically generatedvtime varying leaks, with a detection accuracy of 99%, almost zero false positive detections and the lowest root mean squared error in leak magnitude estimation (0.065 l/s).