Why does modified 99 line topology optimization code outputs blurry images?

Hello,

So, I modified the 99 line topology optimization code found here , so that it has the following boundary conditions (see image below). I'm trying to create topologies for the seat of a chair, so I applied a continuous downwards force of -1 (weight of person sitting on seat) and two fixed supports at the bottom leftmost/rightmost nodes (legs of the chair). I also tried a continuous fixed support (length of the continuous load). However, the resulting images are all very blurry, as in the image below. I'd like to understand why these images are blurry as opposed to the ones found in the 99 line topology optimization article (where the boudnary conditions are for cantilever beams). Unfortunatley, I can't post the code but any insight/guesses would be appreciated. Also, I did try varying "rmin" and "penal" to see if this made the image less blurry but did not help.

[haisen]

It seems an orignal problem raised by SIMP method.

In SIMP, the filter size and the penalty item is related to this problem.

An Improved Method for Decreasing Intermediate Density Elements in Topology Optimization

Please check this paper which may improve this problem.