改进风驱动优化算法在水库调度中的应用

风驱动算法是一种新兴的基于群体迭代启发式的全局优化算法，与遗传算法、布谷鸟算法等相比，具有明确的物理背景，但该算法避免不了易陷入早熟和收敛效率慢的问题。针对早熟，本文提出了扰动策略，对当前最优适应度值对应的任一元素进行扰动，且随着迭代次数的增加，扰动量逐渐减少。针对收敛效率不高，提出了空间压缩策略，采用奇偶相间的方式，通过计算约束更新解的上下限以保证该解是可行解。将改进的风驱动优化算法运用到某水库的优化调度中，并与粒子群算法和标准风驱动算法进行比较。结果表明改进的风驱动优化算法更为可靠、高效，能以较快速度收敛于最优解，且最优解值更大，为水库优化调度模型求解提出新的解决方案。

Application of an improved wind driven optimization algorithm in reservoir operation

Wind-driven algorithm is an emerging global optimization algorithm based on group iterative heuristics. Compared with genetic algorithm and cuckoo algorithm, it has a clear physical background, but the algorithm still cannot avoid the problems of prematurity and slow convergence efficiency. Aiming at prematurity, this paper proposes a perturbation strategy to perturb any element corresponding to the current optimal fitness value,and as the number of iterations increases, the amount of perturbation should gradually decrease. Aiming at the low convergence efficiency, a space compression strategy is proposed, which adopts the method of odd and even phases,and constrains the upper and lower limits of the solution through calculation to ensure that the solution is feasible. The improved wind-driven optimization algorithm is applied to the optimal scheduling of a certain reservoir and compared with the particle swarm optimization algorithm,as well as the standard wind-driven algorithm. The results show that the improved wind-driven optimization algorithm is more reliable and efficient. It can converge to the optimal solution faster and has a larger optimal solution value. It proposes a new solution for the optimization of the reservoir operation model.