Tomographic inversion based on evolutionary algorithms for environmental monitoring applications

Electrical impedance tomography (EIT) is a promising monitoring tool for a rapid and fairly economic mapping of underground pollution in soils. It requires a measuring software capable to recover the conductivity distribution inside the region to be monitored starting from direct measurements of power dissipated or difference potential between couples of measurement points, during current injection between pairs of selected electrodes, placed around the prospected soil. In this paper, after a brief description of the EIT principles and the monitoring process, we proceed to a comparative analysis between genetic and more traditional algorithms in terms of their relative metrological performances. The comparison is handled on the basis of laboratories experiences conducted in a controlled conductivity environment in which the objective is the detection of the magnitude and the location of a conductivity anomaly.