Experimental and computational study of the influence of pre-damage patterns in unreinforced masonry crack propagation due to induced, repeated earthquakes

More Info
expand_more

Abstract

Induced seismicity in the north of the Netherlands has recently exposed unprepared, unreinforced masonry structures to considerable earthquake risk. While the ultimate-limit state capacity of the structures is vital to assess the individual’s risk, their behavior during more frequent, lighter earthquakes, leading to ‘lighter damage’, has shown to be strongly linked to economic losses and societal unrest. When observing the light damage caused by minor earthquakes, the existing state of the structure appears to be highly relevant for the final damage intensity and configuration: earthquakes that may have otherwise caused no apparent damage, may intensify existing damage. In particular, incipient damage due to settlements is common in the baked-clay and calcium-silicate brick masonry structures of the region.
This paper details the study of full-scale laboratory walls, pre-damaged following typical (crack) patterns caused by settlements and tested with quasi-static lateral loads. The aggravation of the damage during a relevant number of load cycles is monitored using full-field digital image correlation. The damage is quantified objectively using a purposely-developed damage parameter.
The tests are used (together with previous studies) to further calibrate computational finite element models, which coupled with detailed soil-structure interaction boundary conditions, are then employed to assess a larger number of structural geometries and pre-damaged configurations exposed to (repeated) induced earthquake acceleration histories.
Both experimental and computational approaches show that settlement pre-damage in masonry structures increases the likelihood and the amount of further damage. This is more easily observed when some initial, yet limited damage exists and the masonry wall is exposed to moderate earthquake vibrations in the order of 30 millimeters per second.

Files

Korswagen_2019_Experimental_an... (pdf)
(pdf | 2.22 Mb)

Download not available