基于改进粒子群优化算法的景观步道设计

The Design of Landscape Trails Based on Improved Particle Swarm Optimization

针对无景观步道的绿化带无法满足行走问题,提出了一种改进的粒子群算法,并在此基础上建立了优化景观步道设计模型。通过引入非线性的动态惯性权重系数,平衡了粒子群算法的全局搜索能力和局部改良能力。仿真结果表明,该算法比常规的无约束优化算法在全局收敛速度、收敛精度及寻优能力等方面有明显的优势。

In view of the problem that green belt without landscape trail or with unreasonable design fails to meet the pedestrian convenient trip, this paper proposed an improved particle swarm optimization algorithm, on which the optimal design model of the landscape footpath was established. By introducing a nonlinear dynamic inertia weight coefficient, the particle swarm algorithm was able to keep its balance of abilities between the global search and local improvement. The simulation results showed that in comparison with the conventional unconstrained optimization algorithm in landscape trail design application, the proposed algorithm had some obvious advantages over it in such aspects as global convergence speed, convergence precision and optimization ability.