1- Department of Civil Eng., Faculty of Eng., Uni. Of Guilan, Rasht ,Iran
2- PhD Student, Uni. of Guilan
Abstract: (4911 Views)
Nowadays, cement is broadly applied to stabilize the soil to improve the mechanical and engineering properties of different soils and to control their deformations and swelling behavior. Nevertheless, due to the high expenses, civil engineers have always been trying to find an economic pozzolanic alternative for cement. In this regard, lots of construction materials such as fly ash, lime, blast furnace ash, pond ash, rice husk ash, etc are added to differentsoil materials to find an optimum replacement for the cement. Among the mentioned materials, rice husk ash (RHA), which is widely available in Guilan, Iran, is an environmentally dangerous material (if dumped in the nature). Hence, application of this material will be both economically and environmentally useful. Also, due to its high applicability, it can be easily applied in civil constructions. Hence, application of RHA in civil/construction projects will be considerably useful. On the other hand, different fibers (e.g. plastic, polyester, polypropylene, etc) are used for engineering purposes to both control the process of crack initiation and increase the material mechanical properties. the length of applied fibers, also, their percentage is two famous controlling parameters of applying fibers in soil stabilization programs. In this paper, a new soil stabilization method is introduced to stabilize Anzali sand using the combination of cement and RHA. Also, the possibility of cement replacement with RHA is investigated. Polypropylene fibers in 0.2 and 0.4 percentages are also added to the samples to control the growth of tensile cracks and to evaluate their effect on the compressive strength of stabilized samples. Hence, cement in 3, 5, 7.5, 10 percentages, also, RHA in 3, 5, 7.5, 10, 12 and 15 percentages are added to the sand samples to increase their compressive strength. Samples are cured for 28 days and unconfined compressive strength tests are conducted on the stabilized and reinforced samples. Based on the test results, compressive strength of all the samples were increase as cement and RHA percentages were increased. Also, RHA is introduced as a capable replacement additive for the cement. In order to make a generalization and provide a relationship for practitioners to use the results of the present study, different techniques can be adopted. Among them, artificial intelligence techniques are most in demanding ones. Neural networks, gen programming languages, genetic algorithms and evolutionary approaches can be applied to provide such relationships. In this paper, evolutionary approaches are considered and using evolutionary polynomial regression technique, simple predictive equation for forecasting UCS is proposed. In this regard, based on the results of conducted un-confined compressive strength tests, Evolutionary Polynomial Regression (EPR) technique is applied and a high accuracy predictive relationship for forecasting UCS of cement-RHA stabilized and polypropylene reinforced sand is presented (coefficient of determination of 94.4%). In addition, sensitivity analysis based on Cosine Amplitude Method (CAM) is carried out to investigate the most and the least effective materials on the compressive strength of samples. CAM analysis showed that although cement and RHA have meaningful effect on the determination of UCS, polypropylene percentage is the most sensitive additive controlling the variation of UCS.
Article Type:
Original Manuscript |
Subject:
Earthquake Received: 2018/01/15 | Accepted: 2024/01/7 | Published: 2019/02/15