Efficient Robust Optimization of Structures Subjected to Earthquake Load and Characterized by Uncertain Bounded System Parameters - Structural Seismic Design Optimization and Earthquake Engineering - page 124

the efficiency of the proposed RDO approach. In
this regard, it is worth mentioning here that one
may fail to identify the complete Pareto front by
the present WSM approach for non-convex Pa-
reto front. This is a potential drawback of WSM
and to overcome this limitation Evolutionary
algorithms like Non-domination Sorting Genetic
Algorithm-II (Deb, 2001) can be applied.

is possible to achieve compare to the conventional
RDO approach is problem dependent. It will spe-
cifically depend on the nature of the performance
function and the entire solution domain, i.e., the
nature of the variation of the objective function
and constraints. This of course needs further
study. The results shown in the numerical example
are for some specific values of the bi-objective
weight factor and penalty factor. However, the
general trend is observed to be similar for other
values. The proposed RDO approach is generic
in nature and can be applied to optimum RDO of
structures under other stochastic dynamic load,
like wind, wave, etc.

SUMMARY AND CONCLUSION

An efficient RDO procedure for structural system
subjected to stochastic earthquake load and char-
acterized by UBB type DVs and DPs is presented.
The associated stochastic constraint is derived
by imposing a limit on the failure probability. To
exclude the complexity of repeated evaluations
of random dynamic responses and their sensi-
tivities (required for RDO solution), the implicit
constraint function is approximated through a
potentially superior MLSM based adaptive RSM
method. The proposed RDO approach improves
the robustness of the performance function using
a new dispersion index which utilizes the weight
factors proportional to the importance of the gra-
dients of the performance function with respect
to uncertain DVs and DPs . The improvement
is achieved in terms of more robustness in the
optimization by allowing more importance to the
uncertain variables which influence the variations
of the performance function and the constraint
violations. Numerical results indicate that the
trends and variations of the optimization results
are in conformity with the well-known two criteria
RDO results. It can be noted that the economy is
not affected in achieving better robustness, as the
proposed RDO approach yields more reduction of
weight for a specified dispersion of the weight.
This implies that the importance factor based RDO
approach captures the optimal design point on a
comparatively flat region than the optimal point
captured by the conventional RDO approach.
However, how much improvement in robustness

REFERENCES

Al-Widyan, K., & Angeles, J. (2005). Model-
based formulation of robust design. Journal of
Applied Mechanics Transaction , 127 (3), 388-396.
doi:10.1115/1.1829728

Ben-Tal, A., & Nemirovski, A. (2002). Robust
optimization-methodology and applications.
Mathematical Programming , 92 (3), 453-480.
doi:10.1007/s101070100286

Bertsimas, D., & Sim, M. (2006). Tractable ap-
proximations to robust conic optimization prob-
lems. Mathematical Programming , 107 (1-2),
5-36. doi:10.1007/s10107-005-0677-1

Beyer, H., & Sendhoff, B. (2007). Robust opti-
mization -A comprehensive survey. Computer
Methods in Applied Mechanics and Engineer-
ing , 196 (33-34), 3190-3218. doi:10.1016/j.
cma.2007.03.003

Bhattacharjya, S., & Chakraborty, S. (2009). Ro-
bust optimization of linear dynamic system with
random parameters under stochastic earthquake
excitation. International Journal of Reliability
Quality and Safety Engineering , 16 (3), 261-279.
doi:10.1142/S0218539309003393