| Examination of the relationship between geomechanical properties of sedimentary and igneous rocks using a collected database from the Ziaran water conveyance tunnel, Iran |
| کد مقاله : 1111-NIEGC2025 |
| نویسندگان |
|
حنان صمدی *1، جعفر حسن پور2 1دانشگاه تهران 2دانشگاه تهران، دانشکدۀ زمین شناسی |
| چکیده مقاله |
| Assessment of rock engineering and geomechanical properties in the pre-construction phase of projects plays a key role in preventing potential problems during tunnel construction. In this study, supervised learning algorithms have been employed to develop models for the prediction of rock geomechanical parameters of sedimentary and igneous rocks using a collected database from the Ziaran water conveyance tunnel (ZrWCT) located in Iran. The factors included water absorption, porosity, tensile strength, slake durability index, specific gravity (apparatus and bulk), elastic modulus, point load index, uniaxial compressive strength, and p-wave and s-wave velocities. Among the data collection during the pre-construction and construction phases of the project, the dataset was classified into seven subsets (zones) based on the different types of sedimentary, volcanic, and igneous rocks (andesite, pyroxenite, tuff, limestone, conglomerate, gabbro, dacite) and the reasonable distribution intervals of the geomechanical factors corresponding to each zone was defined. The relationship between the selected sets of parameters has been investigated to introduce new equations based on the automatic boosting overfit prevention criterion (ABOPC) method.Furthermore, hybrid-optimized machine learning algorithms called AO-TSF, SCG-FFNN, JSO-XGBoost, RF-TSA, HBA-KNN, and TSA-XGBoost have been developed to assess and predict the rock geomechanical properties. The results show that the predicted values are in good agreement with the measured data (most of them have an 80% accuracy rate and MAPE < 0.003). The proposed models can be considered for use in future rock engineering projects, especially hard rock tunnelling with the same lithological types and engineering geological zones defined in this study. |
| کلیدواژه ها |
| ROCK MECHANICS, GEOMECHANICAL PARAMETERS, SUPERVISED LEARNING, COMPUTING MACHINERY TECHNIQUES |
| وضعیت: پذیرفته شده برای ارسال فایل های ارائه پوستر |