Olivier Meyer

MICROWAVE CHARACTERIZATION OF DIELECTRIC MATERIALS USING BAYESIAN NEURAL NETWORKS

This paper shows the efficiency of neural networks (NN), coupled with the finite element method (FEM), to evaluate the broad- band properties of dielectric materials. A characterization protocol is built to characterize dielectric materials and NN are used in order to provide the estimated permittivity. The FEM is used to create the data set required to train the NN. A method based on Bayesian regularization ensures a good generalization capability of the NN. It is shown that NN can determine the permittivity of materials with a high accuracy and that the Bayesian regularization greatly simplifies their implementation.

Direct and Inverse Modeling of a Microwave Sensor Determining the Proportion of Fluids in a Pipeline

A microwave sensor constituted of both low- and high-frequency probes is presented. The role of the sensor is to determine the proportion of mixed fluids in a pipeline used in petroleum engineering to extract oil from an oil layer. In this paper, the annular flow regime in the pipeline is considered. Since the analytical solution for this configuration is not available for high frequencies, a combination of numerical modeling and neural networks (NNs) is used. Two inversion configurations are compared. Good results are obtained with the better one.

Coupled thermal-electromagnetic simulation of a microwave curing cell

The paper presents a practical electromagnetic-thermal simulation of microwave heating during curing process of polymer resin. The model is based on finite elements developed in case of axisymmetry. The finite-element formulation of the electromagnetic problem uses the magnetic field and is well suited to axial symmetry. This model allows to give new insights to understand the differences between microwave and conventional heating.