Affiliation:
1. Institute of Mechanics, Chinese Academy of Sciences
2. University of Chinese Academy of Sciences
3. Aerospace Information Research Institute, Chinese Academy of Sciences
Abstract
Imaging-type FTIR devices provide numerous benefits for the detection and alarm of hazardous gases. This paper presents an improved algorithm for reconstructing the 3D concentration field of gas clouds, utilizing hypothesis testing and a synchronized algebraic iteration algorithm. Specifically designed for use with imaging-type FTIR devices, this algorithm enables rapid reconstruction of gas cloud concentration fields. Using CFD software, an open-space detection scenario for HFC-152a gas was simulated, and the 3D concentration field was reconstructed from dual-angle column concentration data. The accuracy was confirmed, with a deviation of less than 4.6% in re-projected column concentrations along the center streamline and a maximum deviation of 8.8% between simulated and reconstructed voxel concentrations. Laboratory experiments further validated the algorithm. Two sets of line-of-sight angles yielded similar average total mass results calculated from the continuously reconstructed concentration field, measuring 7285.8 mg and 7310.1 mg, with relative standard deviations of 2.4% and 2.7%, respectively. In an open field, an experimental detection of HFC-152a gas leakage was conducted. The algorithm employed facilitated the 3D reconstruction and precise localization of the gas leak source, which underscores the algorithm’s versatility across various environmental contexts and its utility in determining the source of gas leaks. The lab and open field experiments share a same temporal resolution of 2.9 seconds. The algorithm proposed in this article effectively expands the practicality of imaging-type FTIR devices for real-time gas leak monitoring applications.