Neural network based multi-criterion optimization image reconstruction technique for imaging two- and three-phase flow systems using electrical capacitance tomography

In this work a new image reconstruction technique for imaging two- and three-phase flows using electrical capacitance tomography (ECT) has been developed. A brief review on reconstruction techniques developed recently for ECT is also given. The reconstruction technique proposed here is based on multi-criterion optimization using an analogue neural network, hereafter referred to as neural network multi-criteria optimization image reconstruction (NN-MOIRT). The reconstruction technique is a combination between a multi-criterion optimization image reconstruction technique for linear tomography, and the so-called linear back projection (LBP) technique commonly used for capacitance tomography. The multi-criterion optimization image reconstruction problem is solved using Hopfield model dynamic neural network computing. For three-component imaging, the single-step sigmoid function in the Hopfield networks is replaced by a double-step sigmoid function, allowing the neural computation to converge to three distinct stable regions in the output space corresponding to the three components, enabling differentiation among the single phases. The technique has been tested on a capacitance data set obtained from simulated measurement as well as experiment using a 12-electrode sensor. The performance of the technique has been compared with other commonly used iterative techniques, i.e. iterative linear back projection technique (ILBP) and the simultaneous image reconstruction technique (SIRT) for two-phase system imaging, and has shown great improvements in the accuracy and the consistency as compared with those techniques. The technique has also shown the capability of three-phase image reconstruction with high accuracy.