Hassan, Nurfarhana (2018) Probe Drilling Based Prediction Of Rock Mass Strength, Natm-4, Pahang-Selangor Raw Water Transfer Tunnel, Hulu Langat, Selangor, Malaysia. Masters thesis, Universiti Sains Malaysia.
|
PDF
Download (294kB) | Preview |
Abstract
Ramalan keadaan tanah dan batuan geologi terowong adalah salah satu aktiviti penting semasa pembinaan terowong. Sebelum penggalian, ciri-ciri geologi dianggarkan untuk mengelakkan masalah sebagai contoh kemasukan air yang berlebihan atau keruntuhan terowong yang boleh menyebabkan kematian. Kebiasaannya, jumbo terowong dengan gerudi pukulan digunakan untuk siasatan penggerudian. Data yang diperoleh daripada kaedah siasatan penggerudian telah diambil di Hulu Langat, Selangor, Malaysia. Perubahan diperhatikan pada ciri penggerudian semasa siasatan dijalankan dan ia memberikan petunjuk kepada kekuatan tanah, kehadiran rongga besar, jenis batu dan jumlah air bawah tanah. Rekod maklumat yang telah diperolehi ditafsirkan dengan menggunakan kelompok algoritma k-means untuk meramal keadaan tanah yang akan datang. Parameter yang diambil kira adalah kelajuan penggerudian dan peratusan batuan penggerudian memandangkan kedua-dua parameter ini memainkan faktor penting dalam keadaan tanah ramalan hadapan. Parameter ini digunakan untuk mengumpul data dan mengklasifikasi kekuatan batuan di mana ia digunakan untuk membezakan kumpulan kekuatan batu. Berdasarkan kelompok algoritma k-means, kelajuan penggerudian, peratusan keadaan batuan penggerudian dan ramalan ketepatan dikira untuk setiap kelas. Kaedah pengkelasan Japanese Highway (JH) digunakan untuk mengelaskan kelas batuan. Batu kelas B ditakrifkan sebagai kualiti batu yang baik, keras dan sedikit terjejas oleh luluhawa manakala kelas batu E ditakrifkan sebagai batuan yang sangat teruk kualiti and terjejas teruk oleh luluhawa. Dengan menggunakan kaedah kelompok algoritma k-means, ramalan ketepatan diperolehi untuk semua batuan kelas B, CI, CII, D dan E masing-masing adalah 76.2%, 60%, 25.8%, 42.3% dan 51.9% manakala ramalan dari penggerudian pengesanan untuk batuan kelas B, CI, CII, D dan E masing-masing adalah 100%, 30.3%, 36.6%, 84.7% dan 85.2%. _______________________________________________________________________________________________________ Probe drilling in tunnelling is carried out to collect information on the rock mass conditions ahead of a tunnel face. Prior to excavation, geological features are thoroughly studied in order to avoid unexpected difficult ground condition, for example excessive water ingress or tunnel collapse that may cause fatality. Usually, the tunnel jumbo with percussive drill is used for a probe drilling campaign. In this study, probe drilling data was collected from drilling activities recorded at Hulu Langat, Selangor, Malaysia. The observed changes on the drilling characteristic of probe drilling carried out provide us indication on the strength of ground, presence of large cavities, type of rock and the volume of groundwater. The information recorded were interpreted using k-means clustering algorithm to predict the ground condition. Parameters considered are drilling speed and percentages of rock chipping as both play important factors in prediction of ground condition ahead. Both parameters were used to cluster the data and classification of rock strength class. Based on k-means clustering algorithm, range of drilling speed, percentage rock chipping and accuracy was determined for each rock class. Japanese Highway (JH) classification method is used to classified the rock class. Rock class B is defined as a good quality rock, hard and fresh slightly affected by weathering whilst rock class E is defined as very poor to extremely poor-quality rocks and considerably weathered rock mass. By using k-means clustering method, the accuracy prediction was obtained for all rock class B, CI, CII, D and E is 76.2%, 60%, 25,8%, 42.3% and 51.9% respectively whereas prediction from probe drilling for rock class B, CI, CII, D and E is 100%, 30.3%, 36.6%, 84.7% and 85.2% respectively.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Full text is available at http://irplus.eng.usm.my:8080/ir_plus/institutionalPublicationPublicView.action?institutionalItemId=4606 |
| Subjects: | T Technology T Technology > TN Mining Engineering. Metallurgy > TN263-271 Mineral deposits. Metallic ore deposits. Prospecting |
| Divisions: | Kampus Kejuruteraan (Engineering Campus) > Pusat Pengajian Kejuruteraan Bahan & Sumber Mineral (School of Material & Mineral Resource Engineering) > Thesis |
| Depositing User: | Mr Mohd Jasnizam Mohd Salleh |
| Date Deposited: | 06 May 2019 02:46 |
| Last Modified: | 06 May 2019 02:46 |
| URI: | http://eprints.usm.my/id/eprint/44266 |
Actions (login required)
 |
View Item |
