文化算法与投影寻踪融合模型在相似流域优选中的应用

研究文化算法(CA)与投影寻踪(PP)融合模型应用于相似流域优选中的可行性和有效性。以12个小河站控制流域优选为例,建立CA-PP相似流域优选模型,并构建差分进化(DE)算法-PP、和声搜索(HS)算法-PP和粒子群优化(PSO)算法-PP作为对比模型,将优选结果与随机分析法、集对分析法、模糊分析法、灰色分析法的优选结果进行比较。结果表明:CA寻优PP目标函数获得的最优值、最劣值、平均值和标准差均优于DE、HS和PSO算法,具有较好的全局极值寻优能力和收敛稳定性能。CAPP模型对相似流域的优选结果与DE-PP、HS-PP和PSO-PP模型,以及随机分析法、集对分析法、模糊分析法、灰色分析法的优选结果相同,但在优选顺序上存在差异。CA-PP模型用于相似流域优选是可行和有效的,可为同类优选提供新的途径和方法。

Application of fusion model of cultural algorithm and projection pursuit mode in optimization of similar watershed

This article studies the feasibility and effectiveness of the fusion model of cultural algorithm (CA) and projection pursuit (PP) in similar watershed optimization . Based on the optimization of the control catchment of the 12 river stations, this article presents a CA-PP similarity basin optimization model and constructs a differential evolution (DE) algorithm-PP, a harmony search (HS) algorithm-PP and a particle swarm optimization (PSO) algorithm-PP as a contrast model, the optimal results are compared with the optimal results of stochastic analysis, set pair analysis, fuzzy analysis and gray analysis. The results show that the optimal value, the worst value, the mean value and the standard deviation obtained by CA-PP are better than those of the DE, HS and PSO algorithms, and they have better global optimal value and convergence stability. The result of CA-PP model is similar to that of DE-PP, HS-PP and PSO-PP, and the results of random analysis, set pair analysis, fuzzy analysis and gray analysis are the same, but there are differences in the optimization sequence. The CA-PP model is feasible and effective for similar watershed optimization, and it can provide new approaches and methods for similar optimization.