| Budget Amount *help |
¥2,900,000 (Direct Cost: ¥2,900,000)
Fiscal Year 2004: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2003: ¥1,800,000 (Direct Cost: ¥1,800,000)
|
|---|
| Research Abstract |
Because of their enormous numbers, extending the service life of building structures designed by an ordinary structural engineer is one of the most important factors for achieving sustainable society. However, the length of service life of a structure is not considered explicitly in current design provisions. In reliability-based limit state design, the performance level of a structure can be controlled by differentiating load and resistance factors considering the length of service and the variability in load intensity and structural capacity. A practical method for evaluating load and resistance factors is proposed in "Recommendation for Limit State Design of Buildings" by AIJ. However, because of the simple approximation of non-lognormal random variables to lognormal, the reference period is fixed as 50 years, and accordingly, the recommendation cannot be properly used for designing a structure with longer service life.
Also a structure could deteriorate due to environmental stressors and repeated loading. Such deterioration may not be a problem for a structure with short service live, but should be considered for a structure with long service life. However, again no existing design code considers the deterioration of a structural performance during its service life.
For the degradation of structural performance, the investigator has introduced the concept of a resistance factor with which a time-dependent reliability problem can be approximated as a time-independent one. In this research, a practical formula for a resistance reduction factor is proposed classifying the cumulative distribution functions according to the classification of parent distributions of extreme value distributions. Also, an improved simple approximation method, which can consider the service life of a structure, for load and resistance factors is proposed using AFOSM directly. The accuracy and its applicability are investigated using numerical examples.
|
