1: \begin{abstract}

2: In this work, we explore the use of operator splitting algorithms

3: for solving regularized structural topology optimization problems.

4: The context is the classical structural design problems (e.g., compliance

5: minimization and compliant mechanism design), parameterized by means

6: of density functions, whose ill-posendess is addressed by introducing

7: a Tikhonov regularization term. The proposed forward-backward splitting

8: algorithm treats the constituent terms of the cost functional separately

9: which allows suitable approximations of the structural objective.

10: We will show that one such approximation, inspired by the optimality

11: criteria algorithm and reciprocal expansions, improves the convergence

12: characteristics and leads to an update scheme that resembles the well-known

13: heuristic sensitivity filtering method. We also discuss a two-metric

14: variant of the splitting algorithm that removes the computational

15: overhead associated with bound constraints on the density field without

16: compromising convergence and quality of optimal solutions. We present

17: several numerical results and investigate the influence of various

18: algorithmic parameters. 

19: 

20: \smallskip{}

21: 

22: 

23: \emph{Keywords:} topology optimization; Tikhonov regularization; forward-backward

24: splitting; two-metric projection; optimality criteria method

25: \end{abstract}

26: