摘要
建立了以多年年平均发电量最大为目标的混合式抽水蓄能电站水库中长期优化调度模型;使用长序列法并以引用流量和抽水时间作为双控制变量的动态规划算法对水位和抽水时间离散,寻找多种离散水位和抽水时间的最优组合,达到优化水库的各时段运行水位和抽水时间的目的。实例结果表明,在增加抽水蓄能机组后,多年年平均发电量从20.63×108 kW.h增加到了23.06×108 kW.h,保证出力增加了约10 MW,整个调度周期内各时段的水位比不抽水时有所提高;优化算法中的抽水时间以每天1 h为步长的离散处理误差在0.4%以内。
A mid-long term optimal reservoir operation model is built for the hybrid pumped storage power station,which takes the maximum mean annual energy generation for multiple years as its objective. The dynamic programming algorithm,which has two control variables：water flow and pumping time,is applied to discretize the long data sequence of water level and pumping time for searching their best combinations for different periods. Case study shows that,with the added pumped storage units,the mean annual energy generation increases from 2 063 million kW-h to 2 306 million kW-h,the firm power increases about 0.01 million kW,the water level is slightly higher in all periods,and the discretization error is less than 0.4 % when the step of discretized pumping time is 1 hour per day.
出处
《电力自动化设备》
EI
CSCD
北大核心
2012年第4期104-108,共5页
Electric Power Automation Equipment
基金
三峡大学科技基金资助项目(KJ2008B012)
三峡大学硕士学位培优基金资助项目(2011PY020)~~
关键词
抽水蓄能电站
优化
动态规划
人工径流
水电
pumped storage power plants
optimization
dynamic programming
manual runoff
hydroelectric power
