基于仿电磁学算法的梯级水电站多目标短期优化调度  

Electromagnetism-like Mechanism Based Multi-objective Short-term Optimization Scheduling for Cascade Hydro Plants

作　　者：吴杰康[1] 郭壮志[1] 

机构地区：[1]广西大学电气工程学院,广西壮族自治区南宁市530004

出　　处：《中国电机工程学报》2010年第31期14-21,共8页Proceedings of the CSEE

基　　金：国家自然科学基金项目(50767001);国家863高技术基金项目(2007AA04Z197);高等学校博士学科点专项科研基金资助项目(20094501110002);广西高校百名中青年学科带头人资助计划项目(RC20060808002);广西壮族自治区教育厅科研项目(200808MS150);广西壮族自治区研究生教育创新计划项目(105930901001,105930904068)~~

摘　　要：在以水定电的调度策略下,建立水电站获得单时段最佳发电效益的水头和发电流量协调条件,以此为基础构建一天内梯级水电站发电量最大和发电流量正的偏差平方和最大的多目标短期优化调度模型,并采用仿电磁学算法(electromagnetism-like mechanism,ELM)对其求解。采用自适应步长、空间缩减和变异等策略来改善仿电磁学算法的性能。通过递归策略将蓄水量、水量平衡等约束转化为由发电流量表示的约束条件,以达到降低变量维数及提高算法效率的目的。以一个8级梯级水电站为例进行仿真,结果表明仿电磁学算法可有效求解具有复杂约束的非线性优化问题,也验证了所建模型可通过抬高水电站平均发电净水头及降低发电流量,实现对发电量和发电用水量之间的协调,提高梯级水电站运行的综合经济性。A novel multi-objective short-term optimization scheduling model was proposed by considering maximum power generation and the coordinability condition. A novel stochastic optimization algorithm called electromagnetism-like mechanism （ELM） was applied to solve the optimization model with complex constraints efficiently. Space reduction, selfadaptive steps and mutation strategy were also employed for improving the algorithm convergence rate. To enhance the performance of the algorithm, the recursion method was applied for reducing the dimensionality of variables. An example calculation for eight hydro-plants proved the effectiveness of the algorithm in solving the large scale optimization problems and the model in water resource distribution and enhancement of synthesis benefit of cascade hydro plants.

关 键 词：梯级水电站 多目标 短期优化调度 仿电磁学算法 

分 类 号：TM73[电气工程—电力系统及自动化]