| 江燕,刘昌明,胡铁松,武夏宁.新安江模型参数优选的改进粒子群算法[J].水利学报,2007,38(10):1200-1206 |
| 新安江模型参数优选的改进粒子群算法 |
| Improved particle swarm optimization for parameter calibration of Xin'anjiang model |
| |
| DOI： |
| 中文关键词: 参数优选 新安江模型 全局优化 粒子群算法 多种群混合进化 |
| 英文关键词: parameter optimization Xin'anjiang model improved particle swarm optimization algorithm (PSO) parallel swamis shuffling evolution |
| 基金项目: |
|
| 摘要点击次数: 2654 |
| 全文下载次数: 284 |
| 中文摘要: |
| 借鉴竞争演化和多种群混合的思想，对粒子群算法(PSO)进行改进，建立并行种群混合进化的粒子群算法(PMSE-PSO)和序列主-从种群混合进行的粒子群算法(SMSE-PSO)。数值模拟结果表明，这两种改进的粒子群算法具有较高的计算效率、较强的自适应性和稳定性。将PMSE-PSO和SMSE-PSO应用于新安江模型的参数优选中，通过与PSO和SCE-UA的比较可以看出，PMSE-PSO和SMSE-PSO不仅具有较好的全局优化性能和稳定性，而且在调用目标函数次数相同的情况下精度较高，是一种有效的新安江模型参数优选方法。 |
| 英文摘要: |
| Two improved particle swarm optimization algorithm (PSO) including the parallel swarms shuffling evolution algorithm (PMSE-PSO) and serial master slaver swarms shuffing evolution algorithm (SMSE-PSO) were established by combining the particle swarm optimization with competitive evolution and concept of complex shuffling.The comparison with traditional PSO and shuffled complex evaluation algorithm (SCE-UA) shows that both these new algorithms can improve the efficiency, self adaptability and stability of computation.The application of these improved algorithms to parameters optimization of Xin'anjiang model shows that both PMSE-PSO and SMSE PSO remarkably improves the computation efficiency and accuracy. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |