梯级电站厂间负荷分配算法研究

Study on load allocation algorithm for cascade hydropower stations

节能减排形势下,为尽可能节约水资源,电网应下达梯级日内96点负荷曲线,由梯级电站根据自身梯级情况进行厂间负荷分配,力求在最小耗水量条件下满足梯级日内负荷需求,该负荷分配面临高维数、多约束问题。本文以电网下达日内负荷为约束条件,以梯级耗水量最小为目标函数,研究厂间负荷分配求解算法。论文提出层次化厂间负荷分配算法模型,建立负荷随机分配算法得到厂间负荷分配初始解,进而提出功率微增逐次寻优算法获得厂间负荷分配最优解,并进行实例验证。研究表明,算法较好地解决了高维数、强约束条件下厂间梯级负荷问题,达到了节约梯级水资源目的。

As a measure of energy saving and emission reduction,the power grid should give 96-points load curve to cascade hydropower stations within a day for maximum water saving.Load allocation to cascaded hydropower stations should be committed to meet minimum water consumption according to its existing load demand,hydrological conditions and reservoir level.Such a load allocation problem is high dimensional with multi-constraints.This paper presents a study of fast hierarchical algorithm for this problem with consideration of load constraints within a day.For optimal solution we propose a successive approximation algorithm of incremental power with initial solution assigned by a random load method.A case study shows that this algorithm gives a good solution to the high dimensional problem of strong constraints and that water-saving allocation schemes for cascade hydropower stations can be achieved.