Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli
Channel connectors are fairly new alternatives to the shear connectors. Due to their complex behavior and lack of valid approaches, the prediction of shear capacity of these shear connectors is very difficult. The conventional push-out tests and simple modelling of these connectors provide limited g...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://studentsrepo.um.edu.my/8576/ http://studentsrepo.um.edu.my/8576/4/Application_of_ANFIS.pdf |
| _version_ | 1848773695236997120 |
|---|---|
| author | Ali, Toghroli |
| author_facet | Ali, Toghroli |
| author_sort | Ali, Toghroli |
| building | UM Research Repository |
| collection | Online Access |
| description | Channel connectors are fairly new alternatives to the shear connectors. Due to their complex behavior and lack of valid approaches, the prediction of shear capacity of these shear connectors is very difficult. The conventional push-out tests and simple modelling of these connectors provide limited guidance in their structural behavior and the results are valid only for the selected testing protocol. Recent advancements in the area of Artificial Intelligence (AI) have made it feasible to utilize the application of these technologies in the construction industry and structural analysis. Particularly, the use of Artificial Neural Networks (ANNs) has been widely accepted as a reliable tool to solve complex problems in an accurate way. The collective behavior of an ANNs is like a human brain that demonstrates the ability to learn, recall and generalize from training patterns or data. A sub-type of ANNs is Adaptive neuro fuzzy inference system (ANFIS). It integrates both neural networks and fuzzy logic principles with a potential to capture the benefits of both in a single framework. This study aims at predicting the shear strength of channel shear connectors in composite beam comprised of steel and concrete sections using ANFIS as a non-linear modelling tool and the classical Linear Regression (LR) as a linear modelling tool. A set of 1200 experimental data is collected till date and used as an input data of the push-out tests and the output data being the corresponding shear strength which were recorded at all loading stages. The results derived from the use of ANFIS and the LR was then compared. The outcome shows that the use of ANFIS produces highly accurate, precise and satisfactory results as compared to LR. Afterwards, ANFIS network was used to determine which parameters are the most influential on shear strength of channel shear connectors. Two output parameters were analysed, namely load per connector and slip of the shear connectors. To assess the shear strength of channel shear connectors, it is desirable to select and analyse factors or parameters that are truly relevant or the most influential to the shear strength estimation and prediction. This procedure is typically called variable selection that corresponds to finding a subset of the full set of recorded variables that exhibits good predictive abilities. Variable searching using the ANFIS network was performed to determine how the selected parameters affect the shear strength. |
| first_indexed | 2025-11-14T13:46:29Z |
| format | Thesis |
| id | um-8576 |
| institution | University Malaya |
| institution_category | Local University |
| last_indexed | 2025-11-14T13:46:29Z |
| publishDate | 2015 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | um-85762019-07-14T20:16:19Z Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli Ali, Toghroli T Technology (General) TA Engineering (General). Civil engineering (General) Channel connectors are fairly new alternatives to the shear connectors. Due to their complex behavior and lack of valid approaches, the prediction of shear capacity of these shear connectors is very difficult. The conventional push-out tests and simple modelling of these connectors provide limited guidance in their structural behavior and the results are valid only for the selected testing protocol. Recent advancements in the area of Artificial Intelligence (AI) have made it feasible to utilize the application of these technologies in the construction industry and structural analysis. Particularly, the use of Artificial Neural Networks (ANNs) has been widely accepted as a reliable tool to solve complex problems in an accurate way. The collective behavior of an ANNs is like a human brain that demonstrates the ability to learn, recall and generalize from training patterns or data. A sub-type of ANNs is Adaptive neuro fuzzy inference system (ANFIS). It integrates both neural networks and fuzzy logic principles with a potential to capture the benefits of both in a single framework. This study aims at predicting the shear strength of channel shear connectors in composite beam comprised of steel and concrete sections using ANFIS as a non-linear modelling tool and the classical Linear Regression (LR) as a linear modelling tool. A set of 1200 experimental data is collected till date and used as an input data of the push-out tests and the output data being the corresponding shear strength which were recorded at all loading stages. The results derived from the use of ANFIS and the LR was then compared. The outcome shows that the use of ANFIS produces highly accurate, precise and satisfactory results as compared to LR. Afterwards, ANFIS network was used to determine which parameters are the most influential on shear strength of channel shear connectors. Two output parameters were analysed, namely load per connector and slip of the shear connectors. To assess the shear strength of channel shear connectors, it is desirable to select and analyse factors or parameters that are truly relevant or the most influential to the shear strength estimation and prediction. This procedure is typically called variable selection that corresponds to finding a subset of the full set of recorded variables that exhibits good predictive abilities. Variable searching using the ANFIS network was performed to determine how the selected parameters affect the shear strength. 2015 Thesis NonPeerReviewed application/pdf http://studentsrepo.um.edu.my/8576/4/Application_of_ANFIS.pdf Ali, Toghroli (2015) Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli. Masters thesis, University of Malaya. http://studentsrepo.um.edu.my/8576/ |
| spellingShingle | T Technology (General) TA Engineering (General). Civil engineering (General) Ali, Toghroli Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title | Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title_full | Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title_fullStr | Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title_full_unstemmed | Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title_short | Applications of the ANFIS and LR models in the prediction of shear connection in composite beams / Ali Toghroli |
| title_sort | applications of the anfis and lr models in the prediction of shear connection in composite beams / ali toghroli |
| topic | T Technology (General) TA Engineering (General). Civil engineering (General) |
| url | http://studentsrepo.um.edu.my/8576/ http://studentsrepo.um.edu.my/8576/4/Application_of_ANFIS.pdf |