A simulation comparison of 3D reconstruction techniques of muography for blast furnace

To monitor industrial processes within blast furnaces (BFs) using muography, this study focuses on resolving the internal cohesive zone-dead man boundary (CDB). This boundary, which separates the furnace’s semi-plastic melting region (cohesive zone) from its stable coke core (dead man), is a key indicator of operational health. To this end, the effectiveness of various reconstruction algorithms and voxel parameters are systematically evaluated. Theoretical results demonstrate that the Limited-memory Broyden–Fletcher–Goldfarb–Shanno method with a voxel size of 0.2 m achieves the best performance in CDB reconstruction. Furthermore, as disclosed in the investigation of the detector placement configuration, at least four detectors are required to clearly visualize CDB. Eventually, discernible CDB is successfully reconstructed from simulated data for both symmetric and asymmetric BF scenarios without image filtering, which validates the potential of muography to monitor BFs.