Nonlinear Assessment and Optimization Methods for Concrete Structures

点击翻译按钮获取中文摘要

The current practice of Canadian structural engineering is based on traditional analysis methods that go back many years and have shown themselves to be safe, economical and efficient. At the same time there are analysis and design methods in Canadian and International Universities that go well beyond this level of understanding and allow structures to be designed and built which have a series of advantages: a) improved safety against unexpected events such as earthquakes, b) lower overall cost for the same, or higher, level of safety, and c) higher efficiency in terms of lower consumption of energy-intensive construction materials. The problem with these advanced methods is that they tend to be slow to operate and complex to use.Over the past decades, there has also been a dramatic increase in computer power and structural engineers are not yet using that in the most efficient way possible. Indeed we are now at the state where it is possible to achieve some of the goals in the first paragraph with computers doing the optimization process "by themselves" with the engineer merely watching the computer act. This form of unguided optimization is of a concern from the perspective of generation of understanding for engineers but it hasn't generally been a problem yet as the analysis tools are often too complex to be used with optimization methods anyway. Given that such tools are now possible, however, there is a research need to develop optimization methods which include the engineer as a fundamental part of the process.To achieve design optimization of this type, it is necessary that the faster methods of analysis be extended to 3D. In my previous NSERC Discovery grant I planned to generate efficient 2D methods of analysis that more accurately can predict building behaviour accounting for the nonlinear aspects of cracking, yielding and crushing of the concrete. This has now been achieved and it is now possible to extend this to 3D models of buildings with slabs and walls. This type of modelling will allow optimization of 3D buildings to go ahead reasonably efficiently While using computers to optimize designs will allow more efficient structures to be designed, it will also be necessary to develop the optimization process in a way that increases the engineer's understanding of the problems rather than decreasing it as "blind" optimization could do. Thus it will be a key component of the proposed guided optimization that human beings remain part of the design process as the skills generated from that are what are needed to diagnose problems in buildings and come up with truly new design options for the future. That is: the design process must itself be designed to ensure that Canadian engineers remain at the forefront of methods and technologies used around the world.