Do search restarts remove the missed 2 mm sub-search in the laser-cut bracket problem?

What it found

A small search-effort experiment (stylised). At the 2 mm sheet gauge, one differential-evolution run missed the narrow feasible region in 1 of 20 seeds; a second independent run or a doubled population missed in 0 of 40 runs each, at about twice the finite-element solves. Results also shift from an even split towards the lighter of two mass groups (to 70–75% with a restart, 85–95% with the larger population), consistent with the heavier group being a local optimum of the search. Zero misses in 20 does not prove a zero miss rate, and the 2 mm region is itself a coarse-mesh effect, so the lower masses are not better physical designs.

Paper, code and results

• README and method
• Experiments and results
• Code
• Tests

More in Simulation-first design

• Are the two mass groups of the 2 mm laser-cut bracket search two separate basins?The two gauge-2 mass groups differ in web width w (about 14.3–14.9 against 15.2–15.6 mm), not fillet radius, and the lighter group matches ragged isolated feasible pockets on the 2.5 mm mesh rather than a second broad basin; stylised, preliminary.
• Do the isolated feasible pockets of the 2 mm laser-cut bracket survive mesh refinement?The isolated feasible pockets of the gauge-2 bracket scan (web width below 15.4 mm) vanish when the mesh is refined from 2.5 to 1.5 mm and the smallest feasible width rises by about 1 mm; stylised, preliminary.
• Does the feasible boundary of the 2 mm laser-cut bracket converge when the mesh is refined to 0.5 mm?On meshes of 1.0 and 0.5 mm no width in the usable 15.0–16.0 mm window of the gauge-2 bracket scan is feasible (one point at 0.75 mm), and the lowest peak stress rises 11–14% from 1.5 to 0.5 mm; negative, stylised, preliminary.
• Does the 1.5 mm bracket clear buckling when the whole L is one 3D body with a flexible corner?The whole 1.5 mm L-bracket as one 3D body clears the buckling floor under the design load (factor 4.01 vs closed-form 3.47) but not with the load reversed (2.25, below 3); sharp corner, ideal sheet, preliminary.
• How wide must the 1.5 mm L-bracket be to clear the buckling floor when the load is reversed?Widening the 1.5 mm L-bracket web from 18.4 to about 22.6 mm lifts the reversed-load buckling factor from 2.27 to the floor of 3, at +5.9 g (+19%) plate mass; buckling only, sharp corner, ideal sheet, preliminary.
• Do the reversed-load buckling widths of the L-bracket change with more element layers through the thickness?More element layers through the thickness (1 to 4) move the L-bracket's reversed-load buckling factor by at most 1.2% and the width needed for the floor of 3 by at most 0.07 mm at 1.5 to 3.0 mm gauges, so the thicker-sheet designs of simdesign-revgauge stand; coarse in-plane mesh, sharp corner, ideal sheet, preliminary.

All 64 Simulation-first design projects →

Other research topics

Proper scoring & elicitation28Effort & peer prediction9Verification games & slashing19Detection & tolerances14Markets as routers13Prediction-market design9Paying contributors8Decentralised training systems23Digital twins & sim-to-real164LLM chip design loops40Robot manufacturing loops53Ambient AI assistant46Self-replicating manufacturing34Simulation-first design64Engineered clothing45Task-specific AI accelerators40Algorithmic chip-design loops42Open-source purpose-built robots44The software-engineering loop43Adversarial human imitation41Open-source residential construction41Autonomous multi-angle capture34Sensing, imaging and 3D31Energy supply and dense energy storage39Robot societies among people31Agentic co-design of low-cost robots27Laundry-hanging robots25Machine learning on qualitative data23Robots on the building site27Civilizations & group psychology1History & human sciences1

Produced with AI assistance (Claude). Results are from the code in this repository and are simulated or stylised unless the abstract says otherwise; nothing here implies affiliation with or endorsement by any lab, researcher or company cited. MIT licensed.