四分叉铸钢节点的衍生式智能设计研究

Research on generative intelligent design of four-branch cast-steel joint

铸钢节点普遍采用基于经验的传统设计方法,越来越难以满足工程实践和智能制造时代发展的需求。利用衍生式设计方法对某实际工程四分叉铸钢节点的合理构型进行了智能设计研究。详细阐述了衍生式设计的基本方法及数学模型,采用衍生式设计智能生成了若干个四分叉铸钢节点,根据受力性能和成本的评估指标选取了代表性衍生节点,并将代表性衍生节点同初始设计节点、拓扑优化节点和焊接空心球节点进行受力性能的对比分析。为验证衍生式设计结果的可制造性,采用3D打印技术制造出代表性衍生节点的缩尺模型。研究结果表明:利用衍生式设计方法自动生成的四分叉铸钢节点形态具有多样性和创新性;衍生节点模型与其他类型节点相比具有更轻的质量和更佳的受力性能,最大可使节点的质量较初始设计节点减轻3.18%,最大等效应力降低24.78%及最大位移减少37.36%。

The traditional experience-based design method is widely used for cast-steel joints,but it is now more and more difficult to meet the needs of engineering practice and the development of intelligent manufacturing era.To this end,a generative design method was used to study the reasonable configuration of a four-branch cast-steel joint in practical engineering.The generative design method and corresponding mathematical model was described in detail,and then several four-branch cast-steel joints are generated by using the generative design.The representative generation joint models were selected according to the evaluation indexes of mechanical properties and costs,and the mechanical properties of the representative generation joints were analyzed and compared with the initial design joint,topology-optimized joint and welded hollow sphere joint.To verify the manufacturability of the model,the scale model of representative generation joints was manufactured using 3D printing technology.The results show that the configuration of four-branch cast-steel joints generated by the generative design are diverse and innovative in the joint forms.Furthermore,the generative joint model has a lighter mass and better load-carrying performance than other types of joints:it can reduce the mass of the joint by 3.18%,the maximum equivalent stress by 24.78%and the maximum displacement by 37.36%compared with the initial design joint.