نوع مقاله : مقاله پژوهشی
موضوعات
عنوان مقاله English
نویسندگان English
In this article, the Grey Wolf Optimization (GWO) algorithm is discussed, which is considered a global optimization technique capable of improving the global search of particles across the entire search space. The Grey Wolf Algorithm is a relatively new algorithm inspired by the hunting behavior of grey wolves and was first introduced by Mirjalili and his colleagues in 2014. This algorithm has been applied in a few cases to geophysical data. The main goal of the Grey Wolf Optimization Algorithm is to optimize objective functions by drawing inspiration from the behavior of wolf packs to reach better and optimal solutions. Therefore, each of the wolves represents a model with dimensions corresponding to the number of model parameters. The parameters of each wolf (model) include amplitude Coefficient (A), Depth (z), Shape Factor (q), and Center of Mass (x0). The designed algorithm is run for 300 iterations with 30 search agents (wolves), and it is tested on the objective function 10 times, taking the average optimal solution provided by the software as the final parameter.
To evaluate the performance of this method, the gravity field of three synthetic models, namely a sphere, a horizontal cylinder, and a vertical cylinder, both with and without the addition of random noise, is analyzed. Frequency domain estimation of the model parameters is used for each of these models. The results show that the proposed algorithm can accurately estimate the model parameters. Subsequently, the Grey Wolf Optimization Algorithm is applied to analyze the gravity field of the Humble salt dome area in the United States. The results for the studied region indicate that the buried object's center of mass is approximately 4.76 kilometers deep, the domain coefficient is 294.25 units, and its approximate shape is calculated to be similar to a sphere with a calculated shape factor of 1.47, which aligns well with previous studies. The advantage of GWO inversion is its ability to fine-tune the parameters quickly, avoid local minima, and estimate the optimal parameter values.
In this study, the Root Mean Square (RMS) statistical measure is used to compare the measured gravity field and the gravity field calculated based on the estimated parameters. The error between the gravity field values of the synthetic models and the values calculated from the optimal parameters obtained by the Grey Wolf Optimization Algorithm is very small, indicating the algorithm's good performance. Furthermore, the sensitivity of this algorithm to various levels of random noise is investigated, and the results indicate the algorithm's stability against random noise.
کلیدواژهها English
