论文

GPT-6 Astra 驱动 AI 智能体优化含 1650 个 CAD 参数的工程部件设计

AI agents can redesign engineering parts with more than a thousand CAD parameters and nonlinear simulations

精选理由

搞结构优化或 CAE 的朋友可以看看这篇:GPT-6 Astra 只用 15 次仿真就在 1650 个 CAD 参数里把支架应变降了 32%,同预算下 Tosca 只改进 10% 就卡住了。它不压缩参数维度,靠 LLM 读每次求解结果来统筹整体改设计,思路和传统优化器完全不同。

arXiv 论文(2611.26003)测试了 LLM 能否在仿真驱动的设计优化中同时满足三个苛刻条件:少量求解次数、上千维设计空间、以及需要整体重新设计而非局部打磨。研究用 GPT-6 Astra 对一个镁支架进行优化,CAD 几何参数化约 1650 个值,仿真预算仅 15 次,每次提案都在 CAD 中建模、划网格并做大变形塑性与接触仿真,LLM 读取每次求解的完整结果。

正文

在 15 次仿真预算内,GPT-6 Astra 将支架弯折处应变高百分位目标降低了 32%;同等预算下工业形状优化器 Tosca 仅改进约十分之一后便停止。其中一个更早的设计在独立的更高保真全展开模型(含压握和损伤)中验证:整体应变比原始设计低 14%,最受损点更晚开始失效,完全展开时损伤少 63%。

对比对象明确:经典贝叶斯或无导数优化器需要人工把设计空间压缩到少数参数,且需要远多于 15 次的求解;Tosca 这类工业优化器每次循环一个求解但只能小幅打磨表面。结果显示通用 LLM 可以将学到的工程知识与仿真数据结合,协调上千个设计值并产出经独立验证的有效设计,对需要非线性有限元仿真的工程团队意味着优化流程可能不再依赖人工降维。

原文 · arXiv cs.AI

AI agents can redesign engineering parts with more than a thousand CAD parameters and nonlinear simulations

Simulation-driven design of real engineering components has to meet three conditions at once. It runs on few solves, since a nonlinear finite-element simulation of a 3D part takes hours; its design space has many values, the hundreds to thousands of values of the part's CAD model; and a better design often needs a redesign, coordinated changes of the part's proportions rather than local polishing of its surface. Automated optimisers meet at most two of these conditions. Classical optimisers, Bayesian or derivative-free, need a person to cut the design down to a few parameters, and on all the values they need far more solves; industrial shape optimisers move every surface node of the mesh in one solve per cycle, but only in small steps that polish rather than redesign. We ask whether a general-purpose large language model (LLM) can meet all three by joining what it knows with what the simulation shows. GPT-6 Astra edits a magnesium stent through a CAD geometry parametrisation of about 1,650 values, with a budget of 15 simulations; every proposal is built in CAD, meshed and simulated with large-deformation plasticity and contact, and the LLM reads the full results of every solve. Within that budget it lowered the objective, a high percentile of the strains in the stent's bends, by 32%; on the same budget, the industrial shape optimiser Tosca improved the original design by a tenth and then stopped improving. An earlier GPT-6 Astra design, 25% lower in the same objective, was checked in an independent, higher-fidelity model of the full deployment with crimping and damage: it was 14% less strained than the original after the full chain, and its most damaged spot began to fail later and carried 63% less damage at full expansion. LLMs can thus join their learned knowledge with simulation data to adjust more than a thousand design values and reach valid designs with real improvements.

产业坐标与信源核验

本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:GPT-6 Astra、Tosca、LLM、仿真优化,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Alexandra Starostina, Alexander Hermann, Mahish Guru, Kartik Bali, Roland Aydin, Christian C. Cyron),经 AITOP 多维评分管线结构化提炼。