TDIP-based petrophysical model to predict permeability of cement-admixed clay | |
| Author | Khin Moh Moh Latt |
| Call Number | AIT Diss. no.GE-16-02 |
| Subject(s) | Engineering geotechnical. Geophysical prediction Petrology |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering in Geotechnical and Earth Resources Engineering with Area of Specialization in Geosystem Exploration and Petroleum Geoengineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. GE-16-02 |
| Abstract | Permeability is one of the most important petrophysical parameters of cement materials. It is also very difficult to test and estimate this property, particularly for the fine-grained soils. The recent advances in geophysics and petrophysics have opened some directions to predict permeability based on resistivity and chargeability data and good results were obtained for sand and sandstones. How to apply similar approach for fine-grained soils remains challenge to practical geophysicists. This method is focused on using the resistivity and time-domain Induced Polarization (TDIP) in combination with geotechnical testing to predict the permeability of cement-admixed Bangkok Soft clay, a material type that is intermediate between coarse-grain and fine-grain soils. The dissertation is structured in five chapters, i.e., introduction, literature review, methodology, results and discussions, conclusions and recommendations. The major works and outputs of the dissertation included: i) laboratory geotechnical and geophysical testing of cement-admixed soft clays; ii) data analysis to correlate geotechnical and geophysical parameters; and iii) proposing the resistivity and TDIP-based model to predict permeability of fine-grained medium and applying them for the cement-admixed Bangkok soft clay. In this study, Bangkok soft clay samples were collected at the depth of 1.0 to 4.0m. The collected clay at 2.0 to 3.0m was selected to be mixed with three different cement contents at water/cement ratio of 1. The sixty three cement-admixed Bangkok soft clay samples were prepared using three different cement contents of 150, 200, 250 kg/m3 for the unconfined compressive strength, oedometer, resistivity and induced polarization tests for 3, 7, 14, 28 and 56 days of curing period. A number of oedometer and unconfined compressive strength tests were performed to investigate the time-dependent development of strength, porosity, and permeability of the cement-mixed clay samples. The oedometer test results showed that the permeability of cement-mixed Bangkok soft clay decreased with the curing time and the higher cement content the lower permeability. The average permeability of cement-admixed Bangkok soft clay based on oedometer test is 6.84×10-12m/s (or) 6.76×10-19 m2. It was found that there were strong correlations between compressive strength; electrical resistivity and curing time. In addition, the geophysical parameters, i.e., electrical resistivity and chargeability, were measured on the same samples. It was observed that the resistivity and chargeability increased with the curing time and good correlations were found with measured geotechnical parameters such as compressive strength, porosity and permeability. By now, some researches showed that permeability of sands and sandstone could be successfully done using resistivity and frequency-domain IP measurements (SIP), but no attempt has been made with time-domain IP data. In this study, a new approach using resistivity and time-domain induced polarization (TDIP) measurements was proposed to predict permeability. Moreover, Fast Fourier Transform (FFT) was analytically performed to transform the time-domain chargeability to frequency-domain chargeability and the permeability was also predicted using resistivity and frequency-domain chargeability model. The permeability predicted by these two models was validated with the directly measured one by traditional oedometer test. The results of predicted permeability using SIP and TDIP were compared and it is found that the permeability prediction using TDIP chargeability had a better coefficient of determination (R2) and average deviation (d) than that using SIP chargeability. Therefore, it can be concluded that the measured total chargeability values from TDIP measurement can be used directly to predict the permeability in the future. As another part of dissertation study, Artificial Neural Network iv (ANN) was applied to predict permeability based on the inputs of (i) only geotechnical data; (ii) only geophysical data; and (iii) combination of geotechnical and geophysical data, i.e., compressive strength, porosity, electrical resistivity and chargeability. The predicted permeability using TDIP-based model and ANN model matched very well with the permeability measured by oedometer tests, with R2 equal to 0.79 and 0.99, respectively. Consequently, the artificial neural network can also be used as a value-added tool to predict permeability of cement-admixed clay. |
| Year | 2017 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. GE-16-02 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Geotechnical Engineering (GE) |
| Chairperson(s) | Pham Huy Giao; |
| Examination Committee(s) | Noppadol Phien-wej ;Pennung Warnitchai; |
| Scholarship Donor(s) | Ministry of Foreign Affairs, Norway; |
| Degree | Thesis (Ph. D.) - Asian Institute of Technology, 2017 |