State Estimation in Process Tomography – Experimental Study in 3D Multi-phase Flow Case

A. Seppänen, A. Peltola, L. Heikkinen, J. Kourunen and J. P. Kaipio

Department of Physics, University of Kuopio P.O.B. 1627, FI-70211 Kuopio, Finland Email: Aku.Seppanen@uku.fi

ABSTRACT

In this paper, state estimation with fluid dynamical models in process tomography is considered. State estimation is a natural choice for image reconstruction scheme in cases of time-varying targets, because the approach allows taking into account the change of the target during the measurements. Further, in cases of process imaging it is possible to incorporate fluid dynamical models into the reconstruction when state estimation approach is used. For experimental validation of state estimation with fluid dynamical models we built a flow loop and carried out EIT (Electrical Impedance Tomography) measurements of a rapidly varying target. The results indicate that the state estimates can be relatively accurate in a case where the flow rate is high and stationary reconstruction methods lead to useless estimates.

Keywords: EIT, state estimation, Kalman filter, fluid dynamical models