双目标优化与生成对抗网络结合的框架结构阻尼器布置方案智能设计方法

为实现框架结构的阻尼器智能化布置，结合减震设计原理和智能算法，采用双目标优化算法和生成对抗网络算法分别进行阻尼器竖向和水平智能布置研究，并将该方法应用到两个框架结构减震设计工程案例中。在框架结构减震设计中，采用双目标优化算法进行阻尼器竖向布置，并与逐层逼近法、工程师设计和非减震设计进行对比，结果表明，采用该优化算法得到的阻尼器竖向布置方案能有效降低层间位移角和楼层加速度，提高结构的抗震性能。在确定各楼层的阻尼器数量后，利用训练好的生成对抗网络生成模型，可快速、自动地选择和确定各楼层阻尼器的平面安装位置，生成的平面布置与工程师设计的平面布置在相似性差异度综合评价指标上小于临界值0.1，说明两者相似度较高，且有利于提高原结构的抗扭能力。将双目标优化算法与生成对抗网络相结合，不仅能满足框架结构的减震性能目标，而且可实现阻尼器布置方案的智能设计，提升减震工程设计效率。

Intelligent design method of damper placement scheme for frame structure combining dual-objective optimization and generative adversarial network

In order to achieve the intelligent placement of dampers in frame structures, the dual-objective optimization algorithm (DOOA) and generative adversarial network (GAN) algorithm are employed for the vertical and horizontal intelligent placement of dampers, respectively, based on the damping design principle and intelligent algorithm. Two seismic design engineering cases of frame structures are applied. In the seismic design of frame structures, dual-objective optimization is adopted for vertical damper placement. Compared with the layer-by-layer approximation method, engineer-designed optimized damper placement schemes, and non-damping design, the vertical arrangement scheme of dampers obtained by the improved optimization algorithm can effectively reduce inter-story drift angles and floor accelerations, and enhance the seismic performance of the original structure. After determining the number of dampers for each floor, the plane installation position of dampers on each floor can be quickly and automatically selected and determined by using the trained generative adversarial network generation model. The comprehensive evaluation index of similarity difference degree between the generated plane layout and the plane layout designed by the engineer is less than the critical value of 0.1, which indicates that the similarity between the two is high, and it is beneficial to improve the torsion resistance of the original structure. The combination of dual-objective optimization and generative adversarial network can meet the seismic performance objectives of frame structures, and enable to achieve of intelligent design of the damper placement scheme, and improve the efficiency of seismic design engineering.