人工智能在结构拓扑优化领域的现状与未来趋势

结构优化,特别是结构拓扑优化,受到学术界和工业界的广泛关注.通过发展不同的拓扑优化算法,实现了众多具有卓越力学、热学和声学等多学科性能的最优拓扑构型创新设计.然而传统的拓扑优化方法在处理大规模的拓扑优化问题的迭代过程中往往需要多次大规模有限元分析,面临巨大计算量的挑战.近年来,以机器学习为代表的人工智能方法的迅猛发展,成为拓扑优化最具有发展前景的新学科方向.通过将人工智能算法与拓扑优化框架结合,使得拓扑优化的效率大幅提高,同时也为实时拓扑优化的实现提供了可能.本文通过回顾近十年来基于机器学习拓扑优化方法研究的一些重要进展,对截止目前的研究现状进行了简要介绍.由于论文篇幅有限,本综述不涉及该领域的全部文献,其综述范围有限,且与作者本人的研究兴趣密切相关.

Current and future trends of artificial intelligence in the field of structural topology optimization

Structural optimization, especially structural topology optimization, has received extensive attention from academia and industry. Through the development of different topology optimization algorithms, many innovative designs of optimal topology configurations with excellent multi-disciplinary performance such as in mechanical science, thermal science, and acoustics have been realized. However, traditional topology optimization methods often require thousands of iterative steps when dealing with large-scale topology optimization problems, and face the challenge of high computational complexity due to large-scale finite element analyses. In recent years with the rapid development of artificial intelligence methods represented by machine learning, artificial intelligence-based topology optimization has become the most promising new research direction. By combining artificial intelligence algorithms with the topology optimization framework, the efficiency of structural topology optimization is greatly improved, and at the same time, it is possible to achieve a real-time topology optimization design. This paper reviews some major advances in the research of topology optimization methods based on machine learning in the past ten years, and briefly introduces the state of the art up to now. Due to the limited space of the paper, this review does not involve the complete literature in this field, and its review scope is limited closely to the author''s own research interests.