Measurement and prediction of dielectric for liquids based artificial nerve network

Effective complex permittivity measurements of liquids are important in microwave engineering, microwave material processing, microwave chemistry, and electrobiology. Artificial neural network computational modules have recently gained recognition as an unconventional and useful tool for microwave technology. Neural networks can be trained to learn the behavior of the effective complex permittivity of the liquids under irradiation of microwave. It can provide a fast and accurate answer to the task when it has learned. In this paper, we present a simple and convenient method for determining the effective complex permittivity. First, we use a resonant coaxial sensor to measure the reflection coefficients, to check its performance, the electromagnetic field distribution near the sensor and the reflection coefficient is calculated employing the frequency dependent finite difference time domain method. Second, we develop an artificial nerve network and enough simulated materials are utilized to train the networks. Finally, the trained network is employed to predict the effective complex permittivity of liquids, and compare the parameter with pure water obtained from Debye's equation. Results are presented and discussed.

[1]  Fang Wang,et al.  Knowledge based neural models for microwave design , 1997, IMS 1997.

[2]  Om P. Gandhi,et al.  A frequency-dependent finite-difference time-domain formulation for general dispersive media , 1993 .

[3]  Fang Wang,et al.  A hierarchical neural network approach to the development of library of neural models for microwave design , 1998, IMS 1998.

[4]  R. Ludwig,et al.  Suspended Ring Resonator for Dielectric Constant Measurement of Foams , 2006, IEEE Microwave and Wireless Components Letters.

[5]  L. Kempel,et al.  Method of moments solution for a printed patch/slot antenna on a thin finite dielectric substrate using the volume Integral equation , 2006, IEEE Transactions on Antennas and Propagation.

[6]  Qi-Jun Zhang,et al.  Neural Networks for Microwave Modeling: Model Development Issues and Nonlinear Modeling Techniques , 2001 .

[7]  R. B. Standler,et al.  A frequency-dependent finite-difference time-domain formulation for dispersive materials , 1990 .

[8]  Fadhel M. Ghannouchi,et al.  A comparative study of four open-ended coaxial probe models for permittivity measurements of lossy dielectric/biological materials at microwave frequencies , 1996 .

[9]  Kama Huang,et al.  Measurement/computation of effective permittivity of dilute solution in saponification reaction , 2003 .

[10]  L. P. Yan,et al.  A Noninvasive Method for Determining Dielectric Properties of Layered Tissues on Human Back , 2007 .

[11]  Fang Wang,et al.  Robust training of microwave neural models , 1999, 1999 IEEE MTT-S International Microwave Symposium Digest (Cat. No.99CH36282).