基于改进粒子群算法的重力坝断面优化研究

发布时间：2018-02-14 14:58

本文关键词： 重力坝断面优化粒子群算法惯性权重　出处：《西北农林科技大学学报(自然科学版)》2017年01期 　论文类型：期刊论文

【摘要】：【目的】将改进粒子群算法用于重力坝断面的优化计算,为重力坝的优化设计提供支持。【方法】针对传统粒子群算法(PSO)中线性递减的惯性权重极易导致算法陷入局部极值的不足,提出一种改进的粒子群算法(Improved PSO),该算法利用三角函数的相关性质改进惯性权重(w)随时间的动态变化模式,以使惯性权重值在算法的初期保持较大取值,然后逐渐递减而在算法的末期保持较小取值,从而提高粒子群算法的全局搜索能力,增强算法的收敛性能。编制基于改进粒子群算法的重力坝断面优化设计计算程序,对某水利枢纽工程的非溢流重力坝断面进行优化计算分析,并与遗传算法和标准粒子群算法的计算结果进行比较。【结果】采用改进粒子群算法得到的非溢流重力坝的最优断面面积为5 147.3 m~2,而采用标准粒子群算法(SPSO)得到的非溢流重力坝的最优断面面积为5 416.5m~2,前者较后者减小9.45%,极大地提高了经济性;采用改进的粒子群算法得到最优解需要计算15步,而采用标准粒子群算法得到最优解需要计算22步,粒子群算法收敛速度提高了31.8%。通过2种算法计算结果的对比,表明改进的粒子群算法不仅能得到更好的优化结果,而且保持了较快的收敛速度。【结论】改进粒子群算法可以用于大型水利工程结构的优化计算与设计。
[Abstract]:[objective] to apply the improved particle swarm optimization algorithm to the optimization calculation of gravity dam section. [methods] aiming at the linear decreasing inertia weight in traditional particle swarm optimization algorithm (PSO), it is easy for the algorithm to fall into the deficiency of local extremum. An improved particle swarm optimization (PSO) algorithm is proposed in this paper. The improved PSO algorithm uses the properties of trigonometric function to improve the dynamic mode of inertia weight with time, so that the inertial weight can keep a large value in the initial stage of the algorithm. Then decreasing gradually and keeping a small value at the end of the algorithm, so as to improve the global search ability of PSO and enhance the convergence performance of PSO. A program for optimal design of gravity dam section based on improved PSO is developed. The section of a non-overflow gravity dam in a water conservancy project is optimized and analyzed. The results are compared with those of genetic algorithm and standard particle swarm optimization algorithm. [results] the optimum cross section area of non-overflow gravity dam obtained by improved particle swarm optimization is 5 147.3 mm2, while that obtained by standard particle swarm optimization algorithm (SPSO) is obtained by using standard particle swarm optimization (SPSO). The optimum cross-section area of non-overflow gravity dam is 5 416.5 mm2, the former decreases 9.45% than the latter, and the economy is greatly improved. Using the improved particle swarm optimization algorithm to get the optimal solution requires 15 steps, while using the standard particle swarm optimization algorithm to get the optimal solution requires 22 steps. The convergence speed of the particle swarm optimization algorithm is improved by 31.8. the comparison of the results of the two algorithms is made. It is shown that the improved particle swarm optimization algorithm can not only obtain better optimization results, but also maintain a faster convergence rate. [conclusion] the improved particle swarm optimization algorithm can be applied to the optimization calculation and design of large water conservancy structures.
【作者单位】：西安理工大学水利水电工程学院;国网新疆电力公司阿克苏供电公司;
【基金】：国家自然科学基金项目(51409207,51309190) 中央财政支持地方高校发展专项(106-5X1205) 陕西省重点学科建设专项(106-00X903)
【分类号】：TV642.3;TP18

【相似文献】