Computed tomography in process engineering

Abstract Computed tomography (CT), as a non-invasive measurement technique for obtaining digital sectional information, has achieved tremendous success in medical imaging and industrial fields. In process engineering, the primary concern is temporal resolution instead of spatial and density resolution, which is the key difference between process CT and traditional CT. To make this technique more suitable for dynamic and flexible measurement, quite a number of studies have been conducted, and various new tomographic techniques were put forward. In this paper, the general principle of assorted CT used in process engineering is formulated from hardware and algorithm two aspects, and state of the art CT technologies are reviewed with an emphasis on key features concerned by researchers in this field. Challenges like the lack of process orientated calibration and validation methods and data analysis algorithms for dynamic measurement are presented, together with discussions of the limitations such as high cost, safety concerns, data deluge and other miscellaneous topics. With the breakthrough of innovative hardware and data-driven approaches like deep learning, CT will undoubtedly play an even bigger role in serving both academia and industry in process engineering.