Shape and topology optimization in electrical impedance tomography via moving morphable components method

This paper addresses the challenge of reconstructing multiphase conductivity distributions using electrical impedance tomography (EIT). The reconstruction method developed in the paper utilizes the moving morphable component (MMC) approach, where the unknown inclusion(s) to be reconstructed is (are) composed of several candidate morphable components. This work introduces a signed distance-based shape and topology description function (STDF) in lieu of the recently developed hyperelliptic STDF in the MMC approach, thereby absolving the requirement of exponent values. The MMC approach uses explicit geometric entities for the morphable components that are controlled by geometric parameters, such as varying thickness, length, and angle. The optimal inclusion shapes are found by optimizing these geometric parameters in STDFs. Numerical simulation and water tank experiments are used to demonstrate the effectiveness of the proposed method.