This work presents a revision of the structural similarity technique developed for the experimental modeling of marine structures subjected to collision, grounding or similar catastrophic events via miniature models with drastic scale reduction. This revision involved basically the inclusion of combined collapse modes to predict the mechanical behavior of structural members and the redefinition of the flow stress range. The revised technique was validated through numerical simulations of the miniature modeling of nine large-scale marine structures’ experiments found in literature and here presented in the form of nine study cases. Each study case evaluates the accumulated effects of scale reduction, thickness distortion and material distortion in the miniature model as part of the similarity technique. In general, a reasonable-to-good correspondence was observed between the force and absorbed energy responses obtained from reference large-scale structures and their miniature models once brought to the same dimensional scale. Discrepancies between structural responses were quantified by evaluating the normalized root mean square error. By these means, most of the study cases presented errors below 12.5% in terms of force response and below 4.5% in terms of absorbed energy response. On the other hand, lower agreement was encountered when reproducing experiments strongly ruled by progressive buckling or crack initiation/propagation together with severe reduction scales. In these cases, better results are achieved when implementing a more accurate material failure model or by moderating the reduction scale.

这项工作提出了结构相似性技术的修订版，该结构相似性技术是通过采用大幅缩减尺度的微型模型对遭受碰撞，着陆或类似灾难性事件的海洋结构进行实验建模而开发的。此修订基本上包括组合的塌陷模式，以预测结构构件的机械行为和重新定义流应力范围。通过对文献中发现的九种大型海洋结构实验的微型模型进行数值模拟的数值模拟，对验证的修订技术进行了验证，并以九种研究案例的形式进行了介绍。每个研究案例都评估缩微模型，缩尺变形和材料变形在累积模型中的累积效果，这是相似性技术的一部分。一般来说，从参考大型结构及其微型模型获得的力和吸收的能量响应之间达到了合理到良好的对应关系，一旦达到相同的尺寸尺度。通过评估归一化均方根误差来量化结构响应之间的差异。通过这些方法，大多数研究案例的力响应误差均低于12.5％，吸收能量响应误差均低于4.5％。另一方面，当复制实验受到渐进屈曲或裂纹萌生/扩展以及严重缩小的尺度的强烈支配时，会遇到较低的一致性。在这些情况下，当实施更精确的材料故障模型或通过减少缩减规模时，可获得更好的结果。