考虑光滑性约束的水库发电调度图优化方法研究

Study on Optimal Reservoir Operation Rule Curves Considering Smooth Constraint

提出一种改进的基于实数编码的加速遗传算法(RAGA),并用于优化水库发电调度图,用廊道约束并结合平滑函数约束限制调度线的波动,分析比较了不同的编码方案和光滑性约束条件下的计算结果。结果表明,仅对上、下基本调度线进行优化,调度线以每个时段一个实数进行编码,以上、下基本调度线计算加大、降低出力线,利用改进的RAGA结合廊道约束可计算光滑、总发电量较大且能充分反映来水过程的调度图,为水库发电调度图求解提供了新途径。

This paper proposes an improved real coding based accelerating genetic algorithm to optimize the reservoir operation rule curves,which applies the interval constraint and smooth function to reduce the fluctuation of operation rule curves.Various encoding schemes and smooth constraint are compared.The Results show that only the upper and lower basic operation rule curve can be optimized to calculate the increased and reduced output rule curves by using improved RSGA method with real encoding storage level of each period for operation rule curve and interval constraint;At the same time,this approach can obtain smooth operation rule curves,which reflect runoff process with more electricity generation.Thus,it provides new tool for reservoir operation.