改进鸟群算法及其在梯级水库优化调度中的应用

发布时间：2018-07-15 16:38

【摘要】：介绍一种新型群智能仿生算法—鸟群算法(BSA),针对BSA算法进化初期种群多样性的不足,提出基于Lévy飞行策略改进的LBSA算法,并通过2个实例对该算法进行验证:1)利用6个不同维度的典型测试函数对LBSA算法进行仿真测试,仿真结果与基本BSA算法、教学优化算法(TLBO)、差分进化算法(DE)、改进粒子群优化算法(IPSO)、粒子群优化算法(PSO)和混合蛙跳算法(SFLA)的仿真结果进行对比分析.2)为进一步验证LBSA算法的有效性,将其与BSA、TLBO等6种算法用于求解某梯级水库中长期优化调度问题.结果表明:LBSA算法寻优精度优于其他6种算法,具有全局探索及空间勘探能力强、求解精度受维度影响较小、运行速度快、求解精度高等优点,适合求解高维多极值复杂优化问题.LBSA算法优化调度发电量为38.357 3亿kW·h,分别比TLBO、IPSO、DE、PSO和SFLA算法的优化调度结果增加发电量0.005 5、0.008 4、0.010 5、1.467 3和2.678 4亿kW·h,具有较好的优化调度效果.本文通过典型测试函数及实际工程验证了LBSA算法的可行性和高效性,为求解复杂高维的梯级水库群优化调度模型提供了一种全新的途径和方法.
[Abstract]:A new swarm intelligent bionic algorithm, bird swarm algorithm (BSA), is introduced in this paper. Aiming at the deficiency of population diversity in the early stage of evolution of BSA algorithm, an improved LBSA algorithm based on L 茅 vy flight strategy is proposed. Two examples are given to verify the algorithm. (1) six typical test functions of different dimensions are used to test the LBSA algorithm. The simulation results are compared with the basic BSA algorithm. The simulation results of teaching optimization algorithm (TLBO), differential evolution algorithm (DE), improved particle swarm optimization algorithm (IPSO), particle swarm optimization algorithm (PSO) and hybrid leapfrog algorithm (SFLA) are compared and analyzed. Six algorithms, such as BSA-TLBO and BSA-TLBO, are used to solve the long-term optimal operation of a cascade reservoir. The results show that the optimization accuracy of the solution is better than that of the other six algorithms. It has the advantages of strong ability of global exploration and space exploration, less influence of dimension, fast running speed and high precision. LBSA algorithm is suitable for solving high dimensional multiextremum complex optimization problem. The optimal scheduling capacity of the algorithm is 3.83573 billion kW / h, which increases the power generation by 0.005 50.008 4 / 0.010 5 / 1 467 3 and 267.84 million kW / h, respectively, compared with the optimal scheduling results of TLBOO / IPSOD / PSO and SFLA, respectively. This paper verifies the feasibility and efficiency of LBSA algorithm through typical test functions and practical projects, and provides a new approach and method for solving the complex and high-dimensional cascade reservoir group optimal operation model.
【作者单位】：云南省文山州水务局;
【基金】：国家水体污染控制与治理科技重大专项(201307102-006-01) 院士工作站建设专项(2015IC013)
【分类号】：TV697.11;TP18

【相似文献】