The Device Chronicle interviews Sean Weiss on gas imaging non-thermal shortwave infrared (SWIR) detection cameras and OTA software updates
Sean Weiss is an experienced VP of Software Engineering at Kuva Systems with a demonstrated history of working in the oil and energy industry. Sean is an expert in software architecture, design, development, deployment, testing, and scaling. He also believes that doing something to reduce methane emissions from industrial processes is worthwhile as we strive help oil and gas companies produce hydrocarbons with minimal impact on the environment
Methane emissions are a huge problem. Methane is the second most abundant anthropogenic GHG after carbon dioxide (CO2), accounting for about 20 percent of global emissions. Methane is more than 25 times as potent as carbon dioxide at trapping heat in the atmosphere and Methane only stays 27 years in the atmosphere; both of these attributes make it a good target for global warming emission reduction; it costs the least to try to tackle climate change by addressing rogue methane emissions from industrial operations. Up to this point, Oil and gas industry methane emissions have proved difficult to monitor , and monitoring technology now exists to do this better at upstream and midstream facilities. Sean says “near Real-time methane monitoring systems are now possible.” He continues “Methane is not detectable in the visible range of the spectrum, so it is hard to track; you can do it by using lasers do detect when it crosses the beam, or using infrared FLIR cameras, showing the leak’s thermal difference compared to the background, or, as Kuva does, analyzing a part of the infrared spectrum reflected from sunlight to detect and quantify Methane molecules in the air.”
Sean believes that the best approach to monitoring for methane emissions is to marry the technologies together and provide actionable information to the oil field operators to comply with the new regulations that are coming . In the US and Canada, handheld thermal inspections are required once per quarter to measure emissions using a thermal camera. A technician must drive to the asset in a remote location, take the methane emission readings and register them. A consortium of companies, including Kuva, have requested a regulation change, which would allow the use of its non-thermal cameras, and other innovations, to comply with the regulations. The industry would install Kuva cameras and could visibly monitor remotely from a control center. There would no longer be the requirement to send a technician out in the field to look for leaks, reducing operational costs. There is good movement in the regulatory aspects. The oil and gas industry is looking for solutions for this proactively.
Sean says Kuva has tried to find the most cost-effective way to solve the problem: in the Kuva camera, there is an infrared sensor that looks at changes in the spectrum, scans the air, and the sunlight bounces into the camera. Using a method known as spectroscopic absorption, It calculates the number of Methane molecules and other Volatile Organic Compounds between the camera and the reflected sunlight. It creates an image and overlays a colorized plume of the gas and how it moves, and identifies its source so that it can be remediated. Sean says the Kuva camera is a disruptive technology for the industry due to its low price point and ease of deployment.
The Kuva solution also eliminates costly false alarms that affect other emissions monitoring technology as well as expedite the leak repair process due to its imaging technology and helps to find the root cause of the leak with Kuva’s emissions data correlated with process data.
Oil and gas companies Marathon and Devon have spoken publicly about using the Kuva technology. Kuva has also successfully raised $11m in series A funding, so the goal now, Sean says, “is to scale and provide the camera at a competitive price.”
The Kuva Gas Imaging Camera is an Industrial IoT device with edge analytics. The methane gas detections are physics-based, but the camera is brilliant and does a lot of the analytics and calculations in the field. Machine learning processes the data, and the imagery creates alarms around accurate signals of gas formation and false positives and improves the detections. Sean says cloud processing is not desirable; local intelligence in a Linux custom board with Linux Yocto OS and Azure IoT Edge is the run time with several applications. The data is transmitted over cellular connectivity or local Ethernet. It is essential to minimize data transfer costs by optimizing processes at the edge. The Azure layer utilizes Mender for updating all applications asynchronously via the Azure IoT Hub.
Mender was easy to set up and implement; it provided an excellent interface for debugging the Kuva cameras and what happens when something goes wrong, for example, when a camera stops panning. Sean says “We were blind before we started using Mender. We (Kuva) had to send technicians to take down the cameras from poles and update them with a USB. Sean concludes “Our engineers cannot see everything around the application layer, so the remote terminal and troubleshooting are very effective for understanding what’s happening with the whole software system in the device.” Sean concludes “100% Over the Air updates are really a game changer for us in terms of ability to improve and fix issues quickly without needing to spend thousands of dollars to send people to troubleshoot in remote locations.”