The effect of directionality and polarity of ground motion on structural response, and evaluation of traditional analysis using principal orthogonal directions.
Ground motion is measured according to the as-installed instrument orientation, and therefore the same ground motion measured by differently oriented instruments will generate different time-based records. New measures of geometric mean have been therefore produced to combat this inconsistency, such as GMRotIpp (Boore et al.) and the maximum rotated measure (ASCE 7-10). However, codes of practice currently state that structures only need to be assessed for two orthogonal directions of input. This project will try to evaluate the validity of this statement by including the effect of directionality and polarity of the ground motion, in order to find out whether codes are conservative or not. This will involve carrying out a series of time-history analysis on different building structures with earthquake ground motion records applied at different directions (angles). The analysis will be carried out using both (GMRotIpp and Maximum Rotated) ground motion measures of directional intensity, to compare their influence on the structural response.
Developing the research question (what is the effect of directionality and polarity of ground motion in structural response?) was basically taking a previous project that was not successfully completed and led to no relevant conclusions.
The project is based on the research started by Damian Grant on the implications of polarity of ground motion for structural design, and the dissertation project started lats year by Graham Mc Williams (UCL). However, it will be more focused on assessing the probability of a set capacity being exceeded as a result of the change in ground motion input, trying to analyse the different response for various structural types.
However, the approach is now more oriented to assess the probability of a set capacity (not a individual member) of being exceeded; therefore the question became wider: how does this change in ground motion input affects the total probability of failure compared to the traditional (orthogonal) analysis?
In order to be able to make clear comparisons, the following structural types have been priorly selected:
The use of these different structural types will help us to compare extreme situations; according to Grant, every real system may be combination of both (one-way and two-ways) systems.
Aims of the Project:
Methodology
1. LITERATURE REVIEW:
based on the references provided by Damian Grant, related to polarity and directionality as a starting point; as well as the previous project developed by Graham Mc Williams. The research question has been outlined based on the original, but has been modified to fit my personal interests. The literature review has started, and the research question could be modified during the process.
Following a post-positivism critical approach, the mere formulation of the research question is subject to change (slight changes) during the research process.
PUBLISHED PAPERS
UNPUBLISHED MANUSCRIPTS
2. SOFTWARE SKILLS.
Two programs will be used for this project: OpenSees and Mathlab. The first one to carry out the non-linear structural analysis, and the second one to help with the iterative process and the display of results. The learning of these two pieces of software will be done from the second until the last week of may. This will include the development of the basic iteration routine for a SDOF structure (circular bridge pier), which will be simply adapted later for the other structural types
3. GROUND MOTION.
The ground motion records will be provided by Damian Grant, selected from the PEER NGA database, and modified using Grants program RSPMatch 2005bi, which takes into account directionality in the process of spectral matching.
4. STRUCTURAL MODELS.
The other five structural models will be developed during the first three weeks of june.
5. STRUCTURAL ANALYSIS STAGE.
This will take place during the last week of June and the first two weeks of July. It consists in response-history analyses for each model and each suite of ground motions, rotatin them 360 degrees at increments of 10 degrees.
6. ANALYSIS OF RESULTS.
Will be carried out during the second half of July, leaving enough time to draw adequate conclusions during August.
Difficulties
