Smoothing inertial method for worst-case robust topology optimization under load uncertainty

Abstract We consider a worst-case robust topology optimization problem under load uncertainty, which can be formulated as minimization of the maximum eigenvalue symmetric matrix. The objective function is nondifferentiable where multiplicity eigenvalues occurs. Nondifferentiability often causes some numerical instabilities in an algorithm such oscillation generated sequence and convergence to non-optimal point. use smoothing method tackle these issues. proposed guaranteed converge point satisfying first-order optimality condition. In addition, it simple thus has low computational cost per iteration even large-scale problem. experiments, we show that suppresses converges faster than other existing methods.