基于改进分解协调的大规模水电站群优化调度  

Large-Scale Hydropower Plants Optimal Scheduling Based on Improved Decomposition-Coordination Method

作　　者：李纯龙 周建中[1] 欧阳硕[1] 吉鹏[1] 

机构地区：[1]华中科技大学水电与数字化工程学院,湖北省武汉市430074

出　　处：《电网技术》2013年第12期3445-3452,共8页Power System Technology

基　　金：国家自然科学基金项目(51239004);高等学校博士学科点专项科研基金资助项目(20100142110012)~~

摘　　要：以梯级总发电量最大为目标建立大规模水电站群发电优化调度模型,对长江流域17座大型骨干性水电站进行长期发电优化调度研究。针对大规模水电站群优化求解中存在的"维数灾"问题,提出改进大系统分解协调优化方法。该方法在综合大系统分解协调和离散微分动态规划优点的基础上,考虑随机生成可行初始解以缩短其生成时间,利用基于相对水头系数的协调因子实现水电站弃水状态的自动修正,提出自适应偏廊道技术以克服子系统寻优效率低下的问题。长江流域17库实例研究结果表明,相比较于传统优化方法,改进方法得到的梯级总发电量提高0.39%,优化时间缩短68.41%;显著缩短优化时间并提高优化效果,能有效解决大规模水电站群优化调度问题。Considering the maximum of total power generation, an optimal generation dispatching model for large-scale hydropower plants is established, which is studied by 17 backbone hydropower plants with regulation ability in the Yangtze River Basin. In order to solve the problem of ＂dimension disaster＂ in large-scale hydropower plants optimal, the paper proposes an improved large-scale system decomposition-coordination optimal method, which composites the advantages of large-scale system decomposition coordinate- on and discrete differential dynamic program, generates viable initial solution randomly to shorten the generation time, exploits coordination factor based on the relative hydraulic head coefficient to correct the state of water discharge automatically, proposes an adaptive bias corridor technology to overcome the problem of inefficient in subsystem optimal. The result of 17 backbone hydropower plants＇ dispatching in the Yangtze River Basin shows that the total generating capacity by the improved method increased by 0.39 %, and optimize time decreased by 68.41%; the method can reduce the optimization time significantly and improve the optimization effect, solve the problem of large-scale hydropower plants optimization scheduling effectively.

关 键 词：大规模水电站群 系统分解协调 自适应偏廊道 初始解 协调因子 

分 类 号：TM721[电气工程—电力系统及自动化]