关键词: 数字孪生, 起重机结构, 闭环优化设计, 应力预测

Abstract: The unreasonable design of crane structure under traditional experience modes leads to mismatch between the bearing capacity and actual service scenarios.For this reason,a closed-loop optimization design method for crane structures based on digital twins was proposed by introducing the digital design mode.The integrated concept of ‘operating status-bearing performance-shape characteristics' was utilized,and a digital twin system for crane structure closed-loop optimization design was constructed from three levels,namely virtual and real operations and performance mapping,data fusion and interaction,service and visualization presentation.For the stress prediction,based on the structural stress point cloud samples under different working conditions,the K-Nearest Neighbor algorithm (KNN),Ball-Tree algorithm and RBF neural network were used to construct a coarse and fine grid node matching model and stress prediction model.The rapid prediction of structural stress under random loads was achieved.Subsequently,for the structural optimization,stress prediction results were employed to construct the constraints on static strength and fatigue strength.Combined with the main control size parameters of the structure and a multi-objective function composed of quality and life matching,an improved NSGA-Ⅱ algorithm based on NDX operator was adopted to complete the closed-loop optimization design of the crane structure.Taking the YDC20/30 lightweight mobile crane as an example,the feasibility and effectiveness of the proposed method and system were verified.

Key words: digital twin, crane structure, closed-loop optimization design, stress prediction