Statistical Prediction Methods for the In-Plane Shear Strength of Partially Grouted Masonry Walls

  • Author / Creator
    Izquierdo Duque, Karren N
  • The behaviour of partially grouted masonry shear walls is complex due to their anisotropic nature and the nonlinear interactions between blocks, mortar, grouted cells, ungrouted cells and steel reinforcement. Currently available code- and research-based shear strength equations give highly variable results when predicting the in-plane shear strength of partially grouted masonry. It is crucial to develop a greater understanding in this area, as sudden shear failures of masonry walls can lead to catastrophic losses of human life and property. This study presents the development of several new in-plane shear strength models for partially grouted masonry walls using stepwise regression and model trees. Stepwise regression allows the number of input variables to be reduced from a large pool of candidates to an appropriate subset for predicting the output variable. It does so by identifying the most significant input variables while eliminating interdependencies between variables that are included in the regression. Model trees split data into groups of similar data so that nonlinear functions can be represented using piecewise linear approximation. In this study, model trees used stepwise regressions as those linear approximations. The models were generated using data compiled from 292 experimentally tested partially grouted masonry shear walls. The data was synthesized and scrutinized to minimize variation between studies and eliminate inconsistencies. The data was also used to test the performance of 17 existing code- and research-based shear strength equations. Both the stepwise regressions and model trees were found to significantly outperform existing code- and research-based shear strength models. The stepwise regressions generally performed slightly better than the model trees, however it was shown that the potential of model trees is significant, and it was suggested that the benefits of model trees may be more apparent with a larger dataset. Five models, including 3 stepwise regression and 2 model trees, were selected as the proposed models. It was found that, of the variables studied, the most significant ones for estimating the shear strength of partially grouted masonry walls are the axial load, wall geometry, compressive strength of mortar, and area of interior vertical reinforcement.

  • Subjects / Keywords
  • Graduation date
    Spring 2021
  • Type of Item
  • Degree
    Master of Science
  • DOI
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.