Gas-leak source identification by inverse problem analysis of infrared measurement data

With the retirement of skilled workers and the aging of gas plant facilities, interest in the development of automated and remote gas detection methods is increasing. Therefore, a gas leak detection method using infrared camera has been proposed. Invisible gas can be detected from infrared absorption of gas and the emitted infrared ray from the gas itself. The piping of a gas plant is very complicated, so it is very useful that the gas leak source can be identified. In this study, the gas leak source identification is performed by the inverse problem analysis using the infrared measurement data. The least residual method is one of the typical inverse problem analysis methods, and it estimates the position of the gas leak source from the sequential infrared images of the gas cloud distribution and flow. It was found from numerical simulation and laboratory experiments that it is possible to estimate the position of the gas leak source even when the leak source is hidden by piping in the infrared image.