摘要: 为提高新安江模型中参数估计的优化精度和算法性能,提出一种改进的人工蜂群(ABC)算法。设计基于最优个体的寻优和保优策略,采用寻优策略提高观察蜂的深度搜索能力,通过保优策略确保侦察蜂不会丢弃当前最优解,从而使算法能够在较短时间内得到收敛。将改进算法应用于新安江模型的参数估计中,并与ABC算法和SCPSO算法的参数估计结果进行对比。实验结果表明,改进算法得到的参数优化精度比ABC算法提高约4%,比SCPSO算法提高约1%,并且具有较快的收敛速度。
关键词:
人工蜂群算法,
新安江模型,
参数估计,
寻优策略,
保优策略,
Nash-Sutcliffe效率系数
Abstract: To improve optimization precision and performance for parameters estimation of Xinanjiang model,the Artificial Bee Colony(ABC) algorithm is introduced and an improved ABC algorithm(BABC) is proposed based on the optimization strategy and reserve strategy of the best individual.In this improved algorithm,optimization strategy is adopted to increase the depth search capabilities of observation bees,and reserve strategy is adopted to ensure scout bees do not discard the current optimal solution,thus the BABC algorithm can converge in a short period of time.The BABC algorithm is applied for parameter estimation of Xinanjiang model and compared with the ABC algorithm and SCPSO algorithm.Experimental results show that the parameters optimization precision of BABC algorithm improves about 4% than ABC algorithm,and improves about 1% than SCPSO algorithm,and BABC algorithm has faster convergence speed.
Key words:
Artificial Bee Colony(ABC) algorithm,
Xinanjiang model,
parameter estimation,
optimization strategy,
reserve strategy,
Nash-Sutcliffe efficiency coefficient
中图分类号:
火久元,张耀南,赵红星. 人工蜂群改进算法及其在参数估计中的应用[J]. 计算机工程, 2014, 40(12): 166-171.
HUO Jiuyuan,ZHANG Yaonan,ZHAO Hongxing. Improved Artificial Bee Colony Algorithm and Its Application in Parameter Estimation[J]. Computer Engineering, 2014, 40(12): 166-171.