Column Laminography: A Phantom Simulation Study for 2D Vertical Cross-Sectional Imaging of a Distillation Column

Abstract

Distillation columns require continuous monitoring to maintain optimal performance, yet conventional gamma-ray column scanning provides only one-dimensional (1D) attenuation profiles, limiting diagnostic capability. This study investigates gamma-ray laminography as an advanced non-destructive technique for reconstructing two-dimensional (2D) vertical cross-sectional images of column internals. Computer simulations were performed using three phantom model which are normal trays, a missing tray, and trays with localized density anomalies. Projection data were reconstructed using Simple Back Projection (SBP) and Filtered Back Projection (FBP) with Ramlak and Cosine filters. Results showed that SBP produced blurred images with artifacts, while FBP methods yielded clearer images with improved contrast; the cosine filter provided the highest image quality. Laminography successfully identified missing trays and localized anomalies that were indistinguishable in 1D profiles, highlighting its superior spatial resolution. These findings demonstrate the potential of gamma-ray laminography to enhance industrial column diagnostics as a complementary method to conventional column scanning, with future work focusing on experimental validation and real-time implementation.