December 2019
For this Company Spotlight, we interviewed Sensoleak Founder and CEO, Shoshi Kaganovsky, about how they are using smart technology to resolve pain points in the oil and gas pipeline sector. Sensoleak provides software solutions that combine artificial intelligence (AI), machine learning (ML), engineering, physics and internet of things (IoT) to accurately detect and prevent critical leaks in pipelines. For more information on Sensoleak, please visit www.sensoleak.com.
Industry Pain Point: Pipelines are considered the safest and most dependable form of transporting oil and gas, but a leak can be detrimental if undetected or falsely detected. Traditionally, the most common leak detection method for pipeline operators is the supervisory control and data acquisition (“SCADA”) system. The SCADA system collects data from sensors that are equipped to the pipeline and relays the information to a control room, where trained operators identify abnormalities. The substantial amount of irrelevant data and reliance on human analysis leads to repeated missed leaks and false alarms.
Failed Technology Introductions: Many new technologies have been introduced to the industry over the years, attempting to “reinvent” operators’ pipeline monitoring operations by installing new sensors and using statistical algorithms to predict failures. None of these technologies have experienced widespread adoption due to poor accuracy, high number of false alarms and the significant costs associated with installing new hardware.
Sensoleak’s Technology Solution: Sensoleak identified the limitations with existing leak detection methods and technologies and developed a software solution that is based on proprietary machine learning algorithms. The algorithms analyze data from the pipeline sensors and provide alerts to the pipeline operators’ control room when a failure is detected. Sensoleak has differentiated its technology by:
Offering a plug-and-play solution that ties directly into operators’ existing SCADA platform without the need for new hardware (i.e. sensors, etc.).
Utilizing machine learning algorithms that account for all variables of a pipeline and how they interact with environmental forces to more accurately identify leaks and provide a holitic view of the pipeline in real-time.
Sensoleak’s technology has proven to be more accurate and operate with a significantly lower rate of false alarms than alternative leak detection technologies. The technology has identified escalating failures up to 50 days in advance of a leak, providing operators with a “predictive” solution that can identify defects prior to a leak occurring.
Closing Thoughts: While Sensoleak is currently focused on penetrating the oil and gas pipeline sector, its technology can be utilized across several industries and applications. Sensoleak can provide predictive failure detection for anything that monitors data in real-time (or close to real-time). This brings up a good question, “Are there technologies currently being used in other industries that can be imminently applied to the oil and gas industry to solve critical pain points?”