مجله ژئوفیزیک ایران

مجله ژئوفیزیک ایران

مقایسه الگوریتم‌های مختلف یادگیری ماشین برای تخمین مقاومت فشاری تک‌محوری خاک رس تثبیت‌شده با سیمان و خاکستر آتشفشانی

نوع مقاله : مقاله پژوهشی‌

نویسندگان
استادیار، دانشکده فنی و مهندسی، دانشگاه اردکان، اردکان، ایران
چکیده
یکی از چالش‌های ساخت‌و‌ساز روی خاک‌های رسی ضعیف، مقاومت فشاری کم و نشست‌پذیری زیاد خاک است؛ ازاین‌رو مهندسان ژئوتکنیک همواره سعی در بهبود مشخصات این نوع خاک با استفاده از روش‌های مختلف تثبیت خاک دارند. برآورد کارایی روش‌های مختلف تثبیت خاک، مستلزم انجام دادن آزمایش‌های متعدد است که فرایندی زمان‌بر و هزینه‌بر است. در این مقاله به بررسی روش‌های مختلف یادگیری ماشین ازجمله رگرسیون خطی، شبکه عصبی پرسپترون چندلایه (MLP)، K نزدیک‌ترین همسایه، تخمین‌گر ماشین بردار پشتیبان (SVR)، درخت تصمیم، جنگل تصادفی و تقویت گرادیان در تخمین میزان مقاومت فشاری تک‌محوری خاک پرداخته می­شود. برای این منظور از نتایج آزمایش نمونه‌های خاک رس تثبیت­شده با استفاده از دو نوع پوزولان‌ در شرایط عمل‌آوری مختلف استفاده شد. متغیرهای مؤثر ورودی الگوریتم‌های یادگیری ماشین شامل زمان و شرایط عمل‌آوری، ملاریته فعال‌ساز قلیایی، نسبت وزنی پوزولان مورد استفاده در خاک رس، نوع پوزولان و نسبت وزنی فعال‌ساز به میزان رطوبت بهینه هستند. جهت بهینه‌سازی و تسریع عملکرد الگوریتم‌ها، تنظیم ابرپارامترها با روش جستجوی مشبک انجام شد. علاوه‌براین به‌منظور کاهش تغییرپذیری نتایج هنگام تقسیم­بندی داده‌ها به مجموعه آموزشی- آزمایشی از روش اعتبارسنجی متقابل ده‌تایی استفاده و عملکرد الگوریتم‌ها با سه معیار متوسط مربعات خطا، ضریب تبیین و میانگین قدرمطلق خطا بررسی شد. با توجه به نتایج بسیار نزدیک به‌کارگیری روش‌های SVR و MLP، این دو روش برای برآورد مقاومت تک‌محوری خاک مورد مطالعه پیشنهاد می‌شوند. همچنین تحلیل حساسیت مدل‌ها بیانگر این است که متغیرهای ورودیِ پوزولان، ملاریته و زمان عمل‌آوری به‌ترتیب بیشترین درجه اهمیت را در پیش‌بینی مقاومت فشاری دارند.
کلیدواژه‌ها

موضوعات


عنوان مقاله English

Comparison of various machine learning algorithms for estimation of uniaxial compressive strength of cement and volcanic ash stabilized clay

نویسندگان English

Khadije Mahmoodi
Hajar Momeni
Assistant Professor, Department of Civil Engineering, Faculty of Engineering, Ardakan University, Ardakan, Iran
چکیده English

One of the challenges in construction on poor clay soils is their low compressive strength and high settlement. Therefore, geotechnical engineers are always trying to improve the characteristics of this type of soil using different soil stabilization methods. Estimation of the efficiency of different soil stabilization methods requires numerous experimental tests, which can be time-consuming and costly. This article examines the performance of various machine learning methods, including linear regression (LR), multilayer perceptron neural network (MLP), K-nearest neighbor, support vector machine regressor (SVR), decision tree, random forest, and gradient boosting machine in estimating the unconfined compressive strength of stabilized clay soil using two types of pozzolans under different curing conditions. The input variables for the machine learning algorithms include time and curing conditions, the molarity of the alkaline activator, the binder replacement, the type of pozzolan, and ratio of the activator to the optimal moisture content. To optimize and speed up the performance of the algorithms, hyperparameter tuning was performed using a grid search method. In addition, to reduce the variability of the results with respect to data division into training and testing sets, a 10-fold cross-validation method was used. The performance of the algorithms was evaluated using three criteria, namely mean square error, coefficient of determination, and mean absolute percentage error. Given the very close results of the SVR method to the MLP method, these algorithms can be confidently recommended for the estimation of UCS of the studied soil. Moreover, based on the performance evaluation criteria, the linear regression method has the worst results among the models used on the experimental data. Also, the sensitivity analysis of the models shows that, in order of preference, the input variables of binder replacement, molarity and curing time have the highest effect on the predicted compressive strength. The results of parametric analysis represent that by increasing the curing time, for both types of binders, namely cement and volcanic ash, uniaxial compressive strength increases. In addition, by increasing the percentage of binders under the optimum water condition, the uniaxial strength of soil increases, while in dry condition with increasing the binder replacement, for volcanic ash stabilized soil, the UCS decreases and for cement stabilized soil it fluctuates. It is observed that the compressive strength has a direct relationship with the amount of molarity. The results indicate also that by increasing the alkali activator/optimum water content up to 1.2, for the curing time of 28 days, the compressive strength increases and then it will be constant.

کلیدواژه‌ها English

Volcanic ash
clay
cement
uniaxial compressive strength
machine learning
Abdullah, H. H., Shahin, M. A., and Sarker, P., 2017, Stabilization of clay with fly-ash geopolymer incorporating GGBFS: Proceedings of the second world congress on civil, structural and environmental engineering (CSEE’17).
Behnood, A., 2018, Soil and clay stabilization with calcium- and non-calcium-based additives: A state-of-the-art review of challenges, approaches and techniques: Transportation Geotechnics, 17, 14-32, doi:https://doi.org/10.1016/j.trgeo.2018.08.002.
Brownlee, J., 2016, Machine Learning Mastery with Python: Understand Your Data, Create Accurate Models, and Work Projects End-to-End: Machine Learning Mastery, San Francisco.
Das, S. K., Samui, P., and Sabat, A. K., 2011, Application of artificial intelligence to maximum dry density and unconfined compressive strength of cement stabilized soil: Geotechnical and Geological Engineering, 29(3), 329-342, doi:10.1007/s10706-010-9379-4.
de Araújo, M. T., Ferrazzo, S. T., Chaves, H. M., da Rocha, C. G., and Consoli, N. C., 2023, Mechanical behavior, mineralogy, and microstructure of alkali-activated wastes-based binder for a clayey soil stabilization: Construction and Building Materials, 362, 129757.
Ghadir, P., and Ranjbar, N., 2018, Clayey soil stabilization using geopolymer and Portland cement: Construction and Building Materials, 188, 361-371.
Ghanizadeh, A. R., Bayat, M., Tavana Amlashi, A., and Rahrovan, M., 2019, Prediction of unconfined compressive strength of clay subgrade soil stabilized with Portland cement and lime using Group Method of Data Handling (GMDH): Journal of Transportation Infrastructure Engineering, 5(1), 77-96.
Ghanizadeh, A. R., Heidarabadizadeh, N., Bayat, M., and Khalifeh, V., 2022, Modeling of unconfined compressive strength and Young's modulus of lime and cement stabilized clayey subgrade soil using Evolutionary Polynomial Regression (EPR): International Journal of Mining and Geo-Engineering, doi:https://doi.org/10.22059/IJMGE.2022.306688.594858
Ghanizadeh, A. R., and Naseralavi, S. S., 2023, Intelligent prediction of unconfined compressive strength and Young's modulus of lean clay stabilized with iron ore mine tailings and hydrated lime using Gaussian process regression: Journal of Soft Computing in Civil Engineering, 7(4).
Ghanizadeh, A. R., and Rahrovan, M., 2019, Modeling of unconfined compressive strength of soil-RAP blend stabilized with Portland cement using multivariate adaptive regression spline: Frontiers of Structural and Civil Engineering, 13, 787-799.
Ghanizadeh, A. R., Rahrovan, M., and Heydarabadi, N., 2021, Modeling of unconfined compressive strength (UCS) of full-depth reclaimed base materials stabilized with Portland cement using Evolutionary Polynomial Regression: Journal of Civil and Environmental Engineering, 51(105), 171-184, doi:10.22059/IJMGE.2022.306688.594858.
Ghorbani, A., and Hasanzadeh Shooiili, H., 2018, Prediction of UCS and CBR of microsilica-lime stabilized sulfate silty sand using ANN and EPR models; application to the deep soil mixing: Soils and Foundations, 58(1), 34-49.
Güllü, H., 2014, Function finding via genetic expression programming for strength and elastic properties of clay treated with bottom ash: Engineering Applications of Artificial Intelligence, 35, 143-157.
Heidari Dezfuli, T., and Ghanizadeh, A. R., 2020, Prediction of compressive and tensile strength of clayey subgrade soil stabilized with Portland cement and iron ore mine tailing using computational intelligence methods: Civil Infrastructure Researches, 6(1), 73-88.
Kaniraj, S. R., and Havanagi, V. G., 1999, Compressive strength of cement stabilized fly ash-soil mixtures: Cement and Concrete Research, 29(5), 673-677.
Miao, S., Shen, Z., Wang, X., Luo, F., Huang, X., and Wei, C., 2017, Stabilization of highly expansive black cotton soils by means of geopolymerization: Journal of Materials in Civil Engineering, 29(10), 04017170.
Miller, H. J., Guthrie, W. S., Crane, R. A., and Smith, B., 2006, Evaluation of cement-stabilized full-depth-recycled base materials for frost and early traffic conditions: Recycled Materials Resource Center, University of New Hampshire, 27.
Miura, N., Horpibulsuk, S., and Nagaraj, T., 2001, Engineering behavior of cement stabilized clay at high water content: Soils and Foundations, 41(5), 33-45.
Mozumder, R. A., Laskar, A. I., and Hussain, M., 2017, Empirical approach for strength prediction of geopolymer stabilized clayey soil using support vector machines: Construction and Building Materials, 132, 412-424.
Raschka, S., Liu, Y. H., and Mirjalili, V., 2022, Machine Learning with PyTorch and Scikit-Learn: Packt Publishing.
Seco, A., Ramírez, F., Miqueleiz, L., and García, B., 2011, Stabilization of expansive soils for use in construction: Applied Clay Science, 51(3), 348-352.
Soleimani, S., Rajaei, S., Jiao, P., Sabz, A., and Soheilinia, S., 2018, New prediction models for unconfined compressive strength of geopolymer stabilized soil using multi-gen genetic programming: Measurement, 113, 99-107.
Sukprasert, S., Hoy, M., Horpibulsuk, S., Arulrajah, A., Rashid, A. S. A., and Nazir, R., 2021, Fly ash based geopolymer stabilisation of silty clay/blast furnace slag for subgrade applications: Road Materials and Pavement Design, 22(2), 357-371.
Theobald, O., 2017, Machine Learning for Absolute Beginners: A Plain English Introduction, 157: Scatterplot Press, London, UK.
Wild, S., Kinuthia, J., Jones, G., and Higgins, D., 1998, Effects of partial substitution of lime with ground granulated blast furnace slag (GGBS) on the strength properties of lime-stabilised sulphate-bearing clay soils: Engineering Geology, 51(1), 37-53.
Xue, X., Yang, X., and Chen, X., 2014, Application of a support vector machine for prediction of slope stability: Science China Technological Sciences, 57(12), 2379-2386, doi:https://doi.org/10.1007/s11431-014-5699-6.
Yi, Y., Zheng, X., Liu, S., and Al-Tabbaa, A., 2015, Comparison of reactive magnesia-and carbide slag-activated ground granulated blastfurnace slag and Portland cement for stabilisation of a natural soil: Applied Clay Science, 111, 21-26.
Yoobanpot, N., Jamsawang, P., and Horpibulsuk, S., 2017, Strength behavior and microstructural characteristics of soft clay stabilized with cement kiln dust and fly ash residue: Applied Clay Science, 141, 146-156, doi:https://doi.org/10.1016/j.clay.2017.02.028.