Neural network approach to ECT inverse problem solving for estimation of gravitational solids flow
Hela Garbaa, Lidia Jackowska-Strumiłło, Krzysztof Grudzień, Andrzej Romanowski
DOI: http://dx.doi.org/10.15439/2014F368
Citation: Proceedings of the 2014 Federated Conference on Computer Science and Information Systems, M. Ganzha, L. Maciaszek, M. Paprzycki (eds). ACSIS, Vol. 2, pages 19–26 (2014)
Abstract. A new method to solve the inverse problem in electrical capacitance tomography is proposed. Our method is based on artificial neural network to estimate the radius of an object present inside a pipeline. This information is useful to predict the distribution of material inside the pipe. The capacitance data used to train and test the neural network is simulated on Matlab using the electrical capacitance tomography toolkit ECTsim. The provided accuracy is promising and shows efficiency to solve the inverse problem in a simple manner and on reduced computational time about 120 times when compared to the existing Landweber iterative algorithm for tomographic image reconstruction that can be encouraging for dynamic industrial applications.