基于改进蝙蝠算法的梯级水库发电优化调度

Optimal Operation of Cascade Hydropower Station Reservoirs Based on the Improved Bat Algorithm

针对梯级水库优化调度高维度、多约束以及非线性优化的特点,将自适应权重以及连续禁忌搜索算法引入标准蝙蝠算法,改善标准蝙蝠算法在水库优化调度应用中出现的早熟收敛且陷入局部最优的问题:一方面利用自适应权重避免因更新步长机制导致寻优能力不足的问题;另一方面利用连续禁忌搜索算法避免因种群多样性差导致陷入局部最优的问题。案例分析结果表明,改进蝙蝠算法能有效运用于水库发电优化调度中,并且与标准蝙蝠算法相比,具有更强全局寻优能力、更高的运行效率,得到的运行调度结果更优。

For the optimal operation of cascade reservoirs having the characteristics of high-dimension,multi-constraint and nonlinear optimization,the adaptive weight and continuous tabu search were introduced into the basic bat algorithm to solve the issue of premature convergence and getting the local optimal result easily in the optimal operation:On the one hand,the adaptive weight was used for avoiding the lack of optimizing ability caused by the step update mechanism;on the other hand,continuous tabu search was used for avoiding getting the local optimal result caused by the decline in population diversity.The case analysis shows that the improved bat algorithm can be effectively applied to the optimal operation of reservoir power generation and compared with the basic bat algorithm,it has stronger global optimization ability,higher operation performance and better operation scheduling scheme.