Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/486797Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Zamri Chik, Prof. Ir. Dr. | |
| dc.contributor.author | Qasim Adnan Aljanabi (P54549) | |
| dc.date.accessioned | 2023-10-11T02:25:37Z | - |
| dc.date.available | 2023-10-11T02:25:37Z | - |
| dc.date.issued | 2014-05-23 | |
| dc.identifier.other | ukmvital:80157 | |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/486797 | - |
| dc.description | Terdapat banyak tanah liat lembut di seluruh dunia termasuklah di lokasi penting di Malaysia. Wujud dua masalah utama apabila menjalankan pembinaan kejuruteraan awam di atas deposit tanah lembut seperti enapan yang berlebihan dan kekuatan ricih yang rendah. Oleh itu, skim pembaikan tanah adalah diperlukan. Pembaikan tanah menggunakan tiang batu (SC) adalah satu kaedah yang berkesan untuk memperbaiki sifat tanah lembut. Permodelan parameter sifat SC adalah penting dalam rekabentuk dan analisis. Pendekatan pemodelan konvensional bergantung kepada dataset yang memerlukan proses yang memakan masa dan banyak input data yang tidak diketahui. Kepintaran buatan, menggunakan struktur matematik yang fleksibel, boleh mengenal pasti hubungan kompleks bukan linear antara data input dan output. Dalam kajian ini, kaedah pembaikan tanah SC telah digunakan dalam projek benteng yang terletak di dua tapak iaitu Kuantan dan Ipoh. Penyelesaian berangka disediakan untuk menilai dan mengira kesan pelbagai saiz zarah tanah kepada sifat tiang batu di bawah benteng menggunakan Kaedah Unsur Terhingga. Hasil keputusan berangka memberikan nilai enapan, tekanan air liang lebihan dan bonjolan sisi bagi tiang batu. Didapati saiz butiran batu hancur dengan geseran dalaman sebanyak 45o mengurangkan jumlah enapan dan bonjolan sisi kepada lebih kurang 1.04, 0.26 peratus daripada tanah liat yang diperkukuhkan dengan batu hancur bergeseran dalaman sebanyak 27.5o. Di samping itu, beberapa kaedah pemodelan telah digunakan; Rangkaian Neural Buatan (ANN), Sistem Inferens Adaptif Neuro-Fuzzy (ANFIS) dan Mesin Sokongan Vektor (SVM). Parameter model sifat SC dipilih dengan teliti berdasarkan kajian terdahulu. Parameter output sifat SC adalah enapan maksimum di permukaan tanah, enapan total maksimum, bonjolan SC, tekanan air liang lebihan pada kedalaman 2 meter dan 4 meter serta faktor penumpuan tegasan. Tiga proses penilaian digunakan untuk menentukan kesan kepada model. Proses penilaian pertama berdasarkan pembahagian pemberat rangkaian neural untuk menentukan kepentingan relatif setiap parameter input dalam rangkaian, manakala proses penilaian kedua dan ketiga menentukan input yang paling berkesan untuk membina model menggunakan parameter tunggal dan pelbagai kemungkinan kombinasi parameter masing-masing. Model SVM disahkan berdasarkan data lapangan dari dua tapak projek tersebut. Model SVM mengatasi semua model yang dicadangkan dan menambah baik ketepatan ramalan untuk semua parameter sifat SC. Model Nu-RBF memberikan prestasi yang lebih tepat daripada model ANN dan ANFIS dengan peningkatan ketara bagi julat ralat antara 6% dan 20% bagi semua parameter SC. Pengesahan terhadap model ini menunjukkan model yang dicadangkan meramal semua parameter dengan baik dengan nilai R2 sama atau lebih tinggi daripada 0.95.,Soft clay soils are fairly widespread all over the world and some of which are in important locations in Malaysia. There are two main problems encountered when undertaking civil engineering constructions in soft soil deposits, such as excessive settlement and low shear strength. Hence, the need of ground improvement schemes is very necessary. Ground improvement by stone columns (SC) is considered to be a very effective method to improve soft soil properties. Modeling the SC parameters is important in any evaluation of design and analysis. Conventional modeling approaches rely on datasets that require time-consuming processes and large amount of unknown input data. Artificial intelligence, a flexible mathematical structure, can identify complex non-linear relationships between input and output data. In this study, the SC soil improvement method has been used in the embankment project located at two sites, Kuantan and Ipoh. A numerical solution was provided in order to evaluate and calculate various effect of grain size on the behavior of stone column under the embankment using Finite Element (FE). The numerical results provide calculated settlement, excess pore water pressure and lateral bulging of the stone column. It was found that used grain size of crush stone with internal friction of 45o reduces the total settlement and lateral bulging to approximately 1.04, 0.26 percent of that of reinforced clay with crushed stone internal friction of 27.5o. In addition, several modeling methods were applied; Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Support Vector Machine (SVM). The parameters of the behavior SC model are selected carefully based on previous studies. The output parameters are maximum settlement at ground surface, maximum total settlement, bulging of SC, excess pore water pressures at depth of 2 meters and 4 meters as well as stress concentration factor. Three evaluation processes were used to determine the effects on the model; The first assessment process was based on partitioning the neural network connection weights in order to determine the relative importance of each input parameter in the network, whereas the second and third assessment processes determined the most effective input to construct the models using a single and various possible combinations of parameters, respectively. SVM model was verified based on field data from the two sites. The SVM model outperformed all the proposed models and improved the accuracy of prediction for all SC parameters. Nu- RBF model has performed more accurately than ANN and ANFIS model with significant improvement ranging of the error from 6% to 20% for all SC parameters. The verification of the proposed model showed that the model satisfactorily predicted all the parameters with R2 values equal or higher than 0.95.,Ph.D. | |
| dc.language.iso | eng | |
| dc.publisher | UKM, Bangi | |
| dc.relation | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | |
| dc.rights | UKM | |
| dc.subject | Embankments | |
| dc.subject | Artificial intelligent methods | |
| dc.subject | Universiti Kebangsaan Malaysia -- Dissertations | |
| dc.title | Behavior of stone column embedded in soft clays under embankment by using finite element and artificial intelligent methods | |
| dc.type | Theses | |
| dc.format.pages | 178 | |
| dc.identifier.callno | TA760.A444 2014 3 tesis | |
| dc.identifier.barcode | 001120 | |
| Appears in Collections: | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_80157+SOURCE1+SOURCE1.0.PDF Restricted Access | 8.83 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.