کمینه سازی اضافه حفاری ناشی از انفجار در تونل ها با استفاده از روش های هوشمند

Abstract

Over break phenomenon in the executive process of a tunneling project is always one of the most important issues. Nowadays, according to the progress of industry and entrance of new technologies to tunneling industry and gradual acceptance, the new methods are replaced instead of traditional methods (drilling and blast). Although, largely the issue of over break has been controlled by project implementers, but the existence of this major issue in tunnel projects has never been eliminated. In this research, using intelligent network of prediction and optimization, over break was discussed. After selecting the best model based on scoring, the selected model was used for optimization. The R2 and RMSE values of the selected model were 0.921, 0.4820, 0.923 and 0.4277 for training and testing, respectively. The Artificial bee colony (ABC) algorithm, which is one of the new optimization algorithms, was used to optimize these parameters of the explosion pattern. Due to the fact that over break is one of the main problems in tunneling, this reduction can contribute to a good tunnel and stability. After creation several models of optimization and variations in its weights, the optimum amount for the over break was 1.63 m2, which is 47% less than the lowest value (3.055 m2). The optimal pattern can be obtained with the least amount of the over break.