基于DE-BP模型隧道围岩的动态分级

针对隧道施工期间的围岩分级问题,根据地质超前预报获得围岩分级指标,提出了基于DE-BP模型的隧道围岩分级方法,并结合VTK技术、三维地质建模方法及数据库技术编写隧道围岩分级软件,将此方法应用于板石隧道的围岩分级中,进行围岩等级可视化显示与施工方案的调整.结果表明:DE-BP模型的均方差明显小于BP神经网络,分级精度显著提高;DE-BP模型围岩分级结果与勘查设计等级基本相同,验证了该模型的合理性,更加适用于隧道围岩动态分级.

Dynamic classification of tunnel surrounding rock based on DE-BP model

Aiming at the problems of surrounding rock classification during tunnel construction period, in terms of the classification indexes of surrounding rock obtained by the geological advanced prediction, a classification method for tunnel surrounding rock based on DE-BP model was proposed. Combined with VTK technology, 3D geological modeling method and database technology, a classification software for tunnel surrounding rock was compiled. The proposed method was used for the surrounding rock classification of Banshi tunnel, and the visual display of surrounding rock grade and the adjustment of construction plan were carried out. The results show that the mean square error of DE-BP model is obviously smaller than that of BP neural network, and the classification accuracy gets significantly improved. The classification results of surrounding rock obtained with the DE-BP model is basically same as that obtained by exploration design, demonstrating the rationality of proposed model. The proposed method is more suitable for the dynamic classification of tunnel surrounding rock.