نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسنده English
In this study, the scour depth around a river bridge pier was predicted using the Gene Expression Programming (GEP) algorithm. First, the parameters affecting scour were identified and then prioritized based on their level of influence through sensitivity analysis. Of the total 176 available data points, 67% were used for training and 33% for testing the model, and an empirical relationship was proposed. The proposed relationship can be used to predict the scour depth around a bridge pier. An examination of the statistical measures, including the correlation coefficient and root mean square error (RMSE), indicates that removing the bridge-pier diameter parameter results in the lowest correlation coefficient and the highest RMSE. Therefore, the bridge-pier diameter is the most important parameter in the proposed model. The results also indicate that removing the critical flow velocity parameter leads to the highest correlation coefficient and the lowest RMSE. Therefore, among the dimensionless parameters, the ratios of median sediment particle diameter to pier diameter (d₅₀/D), flow depth to pier diameter (Y/D), and flow velocity to critical flow velocity (U/Uc) have the greatest to the least influence, respectively. The critical flow velocity has the least influence on the proposed model. After pier diameter, the parameters of flow velocity, upstream flow depth, mean bed-particle diameter, and critical flow velocity have the greatest to the least influence, respectively.
کلیدواژهها English