Electrical capacitance tomography for flow imaging: system model for development of image reconstruction algorithms and design of primary sensors

A software tool that facilitates the development of image reconstruction algorithms, and the design of optimal capacitance sensors for a capacitance-based 12-electrode tomographic flow imaging system are described. The core of this software tool is the finite element (FE) model of the sensor, which is implemented in OCCAM-2 language and run on the Inmos T800 transputers. Using the system model, the in-depth study of the capacitance sensing fields and the generation of flow model data are made possible, which assists, in a systematic approach, the design of an improved image-reconstruction algorithm. This algorithm is implemented on a network of transputers to achieve a real-time performance. It is found that the selection of the geometric parameters of a 12-electrode sensor has significant effects on the sensitivity distributions of the capacitance fields and on the linearity of the capacitance data. As a consequence, the fidelity of the reconstructed images are affected. Optimal sensor designs can, therefore, be provided, by accommodating these effects.