Saeedpanah, I. and Jabbari, E., 2007. Simulation of water pollution in coastal areas using combination of the Meshless Local Petrov-Galerkin scheme with radial basis functions and the fuzzy parameters. Journal of Coastal Research, SI 50 (Proceedings of the 9th International Coastal Symposium), 900 – 906. Gold Coast, Australia, ISSN 0749.0208

Randomly-changing values in time and space of different parameters such as pollutant load sources and physico-chemical coefficients is the difficulty in water with space and time is a basic problem in modeling water pollution modeling. Deterministic modeling is not adequate because it introduces crisp values of these parameters. To tackle the problem, fuzzy set theory is applied to define imprecise parameters which are used in a water pollution model. Physico-chemical coefficients and pollutant load sources are expressed in the form of fuzzy numbers. A two dimensional Meshless Local Petrov-Galerkin (MLPG) with multiquadric radial basis function (MQ-RBF) as the interpolation method combined with fuzzy arithmetic is used for the solution of the advection-diffusion equation for BOD (Biochemical Oxygen Demand) parameter. An upwind scheme is also used in order to reduce the numerical diffusion due to the convective term. The advantage of this method to the common methods in this field is that with the minimum amount of input data, one would be able to produce the necessary output due to the possibility of inputting the implicit and indefinite data in the form of fuzzy values and receiving the output in the same form, this method could replace the statistical methods. Another advantage of this method is that less input data and consequently less computation is needed. So the use of this method, in the cases that the data is not sufficient for statistical methods, could be very much successful so that the ambiguous and accidental parameters could be defined and described in the form of fuzzy data and relations.