Image Reconstruction of an Electrical Capacitance Tomography System Using an Artificial Neural Network

T. D. Suna,c, R. Muddeb, J.C. Schoutena, B. Scarletta, C.M. van den Bleeka

aDepartment of Chemical Engineering, Delft University of Technology, The Netherlands
bDepartment of Transport Phenomena, Delft University of Technology, The Netherlands
cResearch Institute of Petroleum Processing, Beijing 100083, China

Abstract - In this paper, an image reconstruction algorithm for an electrical capacitance tomography (ECT) system is presented. It is based on an artificial neural network (ANN) Four kinds of grid in the same pipe and with different resolutions are discussed. The sensitivity of a pixel in each grid has been calculated and analyzed. The trained neural network has been applied to experimental data for image reconstruction. Through the cases investigated, a practical image reconstruction strategy for a fluidized bed is suggested.

Key words: ANN, ECT, Image Reconstruction, Fluidized Bed.