摘要
针对粒子群算法局部搜索能力较弱和存在早熟收敛的问题,为了有效地控制粒子群算法的全局搜索和局部搜索,提出了将线性递减权重引入到粒子群优化算法中.该算法是从随机解出发,通过追随当前搜索到的最优值来寻找全局最优解,增加了粒子群算法的局部搜索能力.将其算法优化投影寻踪模型,以此构建了线性递减权重粒子群优化投影寻踪模型,将该模型应用到土坝护坡模式优化评价中,选取9个指标作为评判因子,提出适合该地区的土坝护坡优化模式.结果表明:线性递减权重粒子群优化投影寻踪模型可以有效地找到最佳投影方向,计算投影值,根据投影指标值的大小可对方案进行优选.利用该模型对土坝护坡模式进行综合评价是切实可行的.该算法以其实现容易、精度高、收敛快等优点,并且在解决实际问题中展示的优越性,在工程优化领域具有广泛的应用前景.
To overcome the lower local search ability and the problem of premature convergence in particle swarm optimization,and efficiently control the global and local search of particle swarm optimization,particle swarm optimization with linearly decreasing weight was proposed. It started from random solutions,which followed the current search for the optimal value to find the global optimum,increasing the local search ability of the algorithm. By optimizing the algorithm with the projection pursuit model,particle swarm with linearly decreasing weight optimized projection pursuit model was established. The model was applied to optimize evaluation of dam slope protection,and 9 indicators were re-selected as evaluation factors; as a result,an area-suitable optimization model of dam slope protection was proposed. The results show that using particle swarm with linearly decreasing weight optimized projection pursuit model can effectively find out the best projection direction,and calculate the projection values. According to the size of the projection index value,the program can be optimized. It is feasible for comprehensive evaluation of dam slope protection mode. The advantage of this algorithm is ease implementation,high accuracy,fast convergence,and its superiority is demonstrated in solving practical problems. It can be widely used in the field of engineering optimization.
出处
《排灌机械工程学报》
EI
CSCD
北大核心
2015年第12期1044-1048,共5页
Journal of Drainage and Irrigation Machinery Engineering
基金
黑龙江省自然科学基金资助项目(E201054)
关键词
粒子群优化算法:线性递减权重
投影寻踪
模式评价
土坝护坡型式
particle swarm optimization
linearly decreasing weight
projection pursuit
model evaluation
earth dam slope protection
