针对钢闸门优化问题难以获得全局最优解这一问题,提出了一种活跃目标粒子群(Active Target Particle Swarm Optimization,APSO)算法用于露顶式平面钢闸门的优化设计。该算法在传统粒子群(Particle Swarm Optimization,PSO)算法种群中设定一个活跃目标个体,并将该个体引入算法的迭代更新机制中,活跃目标个体的引入增强了算法跳出局部极值、进行全局寻优的能力;并且采用一个综合学习因子代替原始算法中的多个学习因子,进一步提高了算法的收敛速度与稳定性。在满足钢结构强度等约束条件下,以钢闸门总重力为目标函数,对主梁、边柱、面板和次梁等结构参数进行寻优。同时依据优化结果,采用ABAQUS建立有限元模型对钢闸门主梁进行强度复核,结果表明,采用活跃目标粒子群算法对露顶式平面钢闸门进行优化设计能够得到更优的各结构尺寸参数,优化后的钢闸门总重力与文献算例相比降低了15.38%,并且强度复核符合容许应力要求。

Addressing the challenge of obtaining global optimal solutions for steel gate optimization problems, this study introduces an active target particle swarm optimization (APSO) algorithm for the optimization design of emersed plane steel gates. APSO incorporates an active target individual into the conventional particle swarm optimization (PSO) population and integrates it into the algorithm’s iterative update mechanism. This enhancement bolsters the algorithm’s capability to escape local optima and enhances its global optimization performance. Furthermore, the APSO algorithm employs a comprehensive learning factor in place of multiple individual learning factors used in traditional PSO, thereby improving the convergence rate and stability. Under constraints imposed by steel structure strength requirements, the APSO algorithm optimizes key structural parameters, including those of the main beam, side columns, panel, and secondary beams, with the objective of minimizing the total weight of the gate. After optimization, finite element analysis (FEA) is conducted using ABAQUS to verify the strength integrity of the main beam based on the optimized design parameters. Findings indicate that the APSO algorithm effectively optimizes the design of emersed plane steel gates, yielding improved structural dimensions. Specifically, the optimized gate design achieves a 15.38% reduction in total weight compared to previous literature examples, while stringent strength checks confirm compliance with allowable stress limits.