MODEL PARAMETER ESTIMATION BY GLOBAL OPTIMIZATION ALGORITHM ACCO COMPLEMENTED BY AN ANN-BASED ERROR ESTIMATOR

In order to properly simulate the natural phenomena using numerical model, model parameters have to be estimated by an appropriate manner. Authors regard the model calibration as global optimization problem and have applied global optimization techniques, such as adaptive cluster covering method (ACCO), genetic algorithm, and so on, to the estimation of model parameters. Here, new approach using ACCO and artificial neural network (ANN) is proposed for the model parameter estimation. ANN works as an error estimator in this proposed model (ANN-ACCO) . From the comparison of results with ACCO and genetic algorithm (GA), it is shown that the optimization by ANN-ACCO is reasonably carried out with better accuracy and stability. Model parameter estimation was also successfully established by ANN-ACCO, then the some degree of its applicability to model calibration were shown.