پیش بینی خصوصیات مخلوط ماسه با HDPE توسط مدل MLR ،EPR و Stepwise

نوع مقاله : یادداشت پژوهشی

نویسندگان

گروه مهندسی عمران، دانشکده فنی، دانشگاه گیلان، رشت

10.22034/ceej.2024.55984.2242

چکیده

امـــروزه با توجه به افــزایش روزافزون پلاستیک­ های ضایعاتی و کمبود مکان دفن آن، محققین در تلاش برای بررسی روش‌های مختلف استفاده مجــدد و بازیافت آن‌ها هستند. یکی از روش‌ها، استفاده مجدد از پلاستیک­ های ضایعاتی در پروژه ­های مهندسی است ولی ترکیب این مواد با خاک به­ طور قابل­ توجـهی بر خواص آن­ها تأثیر می‌گــذارد. مطالعـه حاضــر به بررســی و مقایسـه روش‌های آماری رگرسیــون خطــی چنـدگانه (MLR) (Multiple Linear Regression، رگرسیون چندجمله­ای تکاملی (EPR) (Evolutionary Polynomial Regression) و رگرسیون گام به گام (Stepwise) به­ منظور پیش ­بینی خصوصیات ژئوتکنیکی مخلوط ماسه- HDPE (Root-Mean-Square Error) خردشده ضایعاتی (پلی­اتیلن با دانسیته بالا) شامل مدول الاستیک، ضریب فشردگی حجمی و ضریب فشار جانبی خاک در حالت سکون (E و  و ) می‌پردازد. داده‌های ورودی از مجموعه‌ای از آزمایشات ادئومتر (Oedometer) بزرگ مقیاس بر روی مخلوط ماسه -HDPE با در نظر گرفتن درصدهای مختلف HDPE (8%، 6، 4، 2 و 0)، دو تراکم نسبی (70% و 40) و سه تنش نرمال مختلف (kPa300، 200 و 10، 0) به ­دست آمده است. عملکرد مدل‌های ارائه شده توسط EPR، MLR و Stepwise، با استفاده از میانگین مربعات خطا (RMSE) (Root-Mean-Square Error)، ضریب تعیین (Artificial Neural Networks) (R2) برای به ­دست آوردن بهترین مدل برازش ارزیابی شد. نتایج نشان داد که مدل EPR و MLR می‌توانند به­ عنوان ابزار قدرتمندی برای مدل‌سازی خصوصیات ژئوتکنیکی ماسه -HDPE و سایر مخلوط­ های با شرایط مشابه استفاده شود. به ­کمک این روش‌ها علاوه بر کاهش هزینه‌ آزمایشات، زمان دسترسی به نتایج آزمایش نیز به طور قابل توجهی تسریع می‌شود. مدل ارائه شده در روش Stepwise برای پیش ­بینی مدل رفتاری مخلوط ماسه -HDPE، نتایج مناسب و دقیقی ارائه نکرده و استفاده از آن پیشنهاد نمی‌شود.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Prediction of Properties of Sand Mixture with HDPE by MLR, EPR and Stepwise Models

نویسندگان [English]

  • Mahyar Arabani
  • Masoomeh Khodabakhshi Seyghalani
Department of Civil Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran
چکیده [English]

The main goal of this research is to present a suitable mathematical model using the MLR (Multiple Linear Regression), EPR (Evolutionary Polynomial Regression), and Stepwise statistical methods. This model aims to predict the relationship between the properties of sand mixed with High Density Polyethylene (HDPE) (as a waste additive material to soil), including the elastic modulus, the coefficient of compressibility, and lateral earth pressure coefficient at rest. The purpose of presenting this model is to reduce laboratory work, and consequently, reduce costs and time. Additionally, the present study focuses on designing and developing a set of models, fitting experimental results, and comparing the best models provided by EPR, MLR, and Stepwise, to find the best modeling approach.

کلیدواژه‌ها [English]

  • Geotechnical properties of sand
  • MLR
  • EPR
  • Stepwise
  • Large-scale oedometer
  • HDPE
Ahangar-Asr A, Faramarzi A, Mottaghifard N, Javadi AA, “Modeling of permeability and compaction characteristics of soils using evolutionary polynomial regression”, Computers & Geosciences, 2011, 37, 1860-1869.
Aroraa S, Keshari A, “Estimation of re-aeration coefficient using MLR for modelling water quality of rivers in urban environment”, Groundwater for Sustainable Development, 2017. https://doi.org/10.1016/j.gsd.2017.11.006
Bruno DE, Barca E, Goncalves RM, de Araujo Queiroz HA, Berardi L, Passarella G, “Linear and evolutionary polynomial regression models to forecast coastal dynamics: Comparison and reliability assessment”, Geomorphology, 2018, 300, 128-140. https://doi.org/10.1016/j.geomorph.2017.10.012
Chakraborty A, Goswami D, “Prediction of slope stability using multiple linear regression (MLR) and artificial neural network (ANN)”, Arab J Geosci, 2017, 10, 385. https://doi.org/10.1007/s12517-017-3167-x
Farooq K, Khalid U, Mujtaba H, “Prediction of compaction characteristics of fine-grained soils using consistency limits”, Arabian Journal for Science and Engineering, 2016, 41 (4), 1319-1328.
Giustolisi O, Savic DA, “Advances in data-driven analyses and modelling using EPR-MOGA”, Journal of Hydroinformatics, 2009, 225-236.
Giustolisi O, Savic, DA, “A Symbolic Data‐driven technique based on evolutionary polynomial regression, Journal of Hydroinformatics”, IWA‐IAHR Publishing, UK, 2006, 3 (8), 207-222.
Karimnader-Shalkouhi S, Karimpour-Fard M, Lashteh Neshaei MA, “Evolutionary Polynomial Regression-Based Models to Estimate Stability of Gravity Hunched Back Quay Walls”, AUT Journal of Civil Engineering, 2018, 2 (1), 79-86.
Karimpour-Fard M, Machado SL, Falamaki A, Carvalho MF, and Tizpa P, “Prediction of Compaction Characteristics of Soils from Index Test’s Results”, Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2019, 43 (1), 231-248. https://doi.org/10.1007/s40996-018-0161-9
Khodabakhshi M, Arabani M, “Statistical investigation on plastic waste recycling by reusing in soil”, Gradevinar 74, 2022, 9, 739-747.
Khodabakhshi M, Arabani M, “Experimental Study on the Static and Dynamic Properties of Silty Sand Mixed with HDPE”, Indian Geotechnical Journal, 2023. https://doi.org/10.1007/s40098-023-00717-4
Khodabakhshi M, Arabani M, “Laboratory investigation of the use of HDPE-sand mixture as a reinforced roadbed”, Transportation Infrastructure Engineering, 2022, 8 (4), 63-78. https://doi.org/10.22075/jtie.2022.29062.1618
Khodabakhshi M, “Investigation of the deformability characteristics of Anzali sand mixed with HDPE, using an oedometer”, PhD Thesis, University of Guilan, Winter, 2012.
Mansourfar K, Advanced Statistical methodes”, Publications of the University of Tehran, 2006.
Merdun H, Çınar Ö, Meral R, Apan M, “Comparison of artificial neural network and regression pedotransfer functions for prediction of soil water retention and saturated hydraulic conductivity”, Soil and Tillage Research, 2006, 90 (1-2), 108-116.
Mukhlisin M, Abd Rahman ASB, “Prediction of Atterberg limits via ANN and ANFIS: a comparison. Proc Int Conf on Environmental Science and Geoscience (ESG ‘14)”, Venice, Ital, 2014, 69-74. http://www.europment.org/library/2014/venice/ENVIR.pdf.
Monjezi M, Aminikhoshalan H, Yazdianvarjani A, “Prediction of y rock and back break in open pit blasting operation: a neurogenetic approach”, Arabian Journal of Geosciences, 2010. https://doi.org/10.1007/s12517-010-0185-3
Omar M, Shanableh A, Mughieda O, Arab M, Zeiada W, AlRuzouq R, “Advanced mathematical models and their comparison to predict compaction properties of fine-grained soils from various physical properties”, Soils and Foundations, 58, 2018 1383-1399. https://doi.org/10.1016/j.sandf.2018.08.004
Rezania M, Javadi AA, Giustolisi O, “Evaluation of liquefaction potential based on CPT results using evolutionary polynomial regression”, Computers and Geotechnics, 2010, 37, 82-92.
Rezania M, Faramarzi A, Javadi AA, “An evolutionary based approach for assessment of earthquake- induced soil liquefaction and lateral displacement”, Engineering Application of Artificial Intelligence, 201124 (1), 142-153.
Rezania M, Javadi AA, Giustolisi O, “An evolutionary based data mining technique for assessment of civil engineering systems”, Journal of Engineering Computations, 2008, 25 (6), 500-517.
Silva AV, Dantas Neto SA, Souza Filho FA, “A simplified method for risk assessment in slope stability analysis of earth dams using fuzzy numbers”, Electronic Journal of Geotechnical Engineering, 2016, 21 (10), 3607-3624. http://www.ejge.com/2016/Ppr2016.0304ma.pdf
Sivrikaya O, “Models of compacted fine-grained soils used as mineral liner for solid waste”, Environmental Geology, 2008, 53, 585-1595. https://doi.org/10.1007/s00254-007-1142-7.
Tadesse AA, “Construction quality monitoring of emnankment dams using pore water pressure the casevod Kesem dam”, Civil and Environmental Engineering (Major in Hydraulic Engineering), Addis Ababa, Addis Ababa University, Degree of Master of Science: 85,2014.