Research Items (5)
- Jan 2018
Solid waste in the form of construction debris is one of the major environmental concerns in the world. Over 20 million tons of construction waste materials are generated in Tehran each year. A large amount of these materials can be recycled and reused as recycled aggregate concrete (RAC) for general construction, pavement and a growing number of other works that drive the demand for RAC. This paper aims to predict RAC compressive strength by using Artificial Neural Network (ANN). The training and testing data for ANN model development were prepared using 139 existing sets of data derived from 14 published literature sources. The developed ANN model uses six input features namely water cement ratio, water absorption, fine aggregate, natural coarse aggregate, recycled coarse aggregate, water-total material ratio. The ANN is modelled in MATLAB and applied to predict the compressive strength of RAC given the foregoing input features. The results indicate that the ANN is an efficient model to be used as a tool in order to predict the compressive strength of RAC which is comprised of different types and sources of recycled aggregates.
Fiber reinforced polymers (FRP) are one of the most commonly used materials for rehabilitation and retrofit of structures. In some cases like the improvement of ancient buildings, these materials do not play a good role and show some defect in their performance. Fiber Reinforced Mortars (FRM) is a new generation of reinforcing materials is invented as external structural and seismic reinforcement. One of the most desirable functions of this type of materials is related to the in-plane shear strength of masonry walls. This paper aims to propose a formula for estimating the shear strength capacity of this kind of walls by using computational intelligence methods, ANN-GMDH was used for this purpose. This approach used 48 experimental dataset results to propose a model to be able to predict the shear strength of FRM strengthened masonry. The proposed model has a correlation coefficient of 0.95, which represents the high efficiency of the model.
- Jul 2017
Fiber reinforced polymers (FRP) are one of the most commonly used materials for rehabilitation and retrofit of structures, mainly concrete structures. There are many reported studies on the behavior of these materials under different types of loading and structures. The aim of this study is to declare the challenges and investigate the efficiency of existing models for samples with rectangular and square sections. Moreover, a new empirical model is developed to predict the strength of FRP confined samples using gene expression programming (GEP). The best model is selected after establishing and controlling several models with different combinations of the influencing parameters. An analysis is carried out to check the performance of the introduced model based on the common criteria such as correlation of coefficient (R). This parameter is 0.923 and 0.922 for training and testing datasets, respectively which reveals a good agreement between the predicted and measured data in the developed GEP model.