A Closed Optical Path Trace H2S Sensing System Based on Vacuum Ultraviolet Differential Absorption Spectroscopy Combined with Reconstructed Domain Point Cloud Segmentation

Accurate detection of trace hydrogen sulfide (H₂S) plays an indispensable role in the fields of industrial emission reduction and power troubleshooting. In view of the oscillatory absorption characteristics of H₂S in the 185-210 nm band, ultraviolet differential optical absorption spectroscopy (UV-DOAS) has been recognized as a promising technique for measuring H₂S concentrations. However, spectral overlap caused by the presence of oxygen (O₂) in the open optical path and in the mixture, along with the irregular absorption characteristics of H₂S, makes it challenging to assess trace H₂S concentrations using the vacuum ultraviolet band. In this study, we propose a closed optical path H₂S sensing system based on vacuum ultraviolet absorption spectroscopy combined with reconstructed domain point cloud segmentation. First, a closed optical path structure is used to eliminate O₂ present in the optical path gaps. Then we present a spectral line analysis method based on sinusoidal spectral reconstruction (SSR) combined with point cloud segmentation. This approach effectively eliminates spectral interference of O₂ on H₂S spectra within the reconstructed domain. Finally, the irregular absorption features of H₂S are organized to facilitate the accurate measurement of H₂S at low concentrations. Results reveal that the sensing system can measure H₂S dynamically at low concentrations using the vacuum ultraviolet band, with a mean relative error (MRE) of 1.61%. Meanwhile, the detection limit for H₂S reached 41.2 ppb at a detection time of 100 s. To the best of our knowledge, this is the best level of H₂S measurement for direct application of UV-DOAS technology.