摘要
我们研究了电力市场的输电阻塞管理,针对目前电力市场中出现的输电阻塞,提出了阻塞费用的计算办法,机组出力分配预案的算法,以及重新调整预案的模型,得到如下结果:问题1:根据32组试验数据,利用多元线性回归建立了6条主要线路的潮流值关于8台机组出力的线性表达式,利用SAS8软件得到回归方程都通过了显著性检验,复相关系数都不低于0.9995,最大均方误差不超过0.03995,相对误差不超过0.0267%,方案0的最大预测误差不超过0.0447%,说明该表达式很好地反映了线路潮流值与发电机组出力的关系。问题2:我们给出了一种合理的计算阻塞费用的规则:序外容量和序内容量都按照预案清算价和新方案出力对应报价之差计算,这在一定程度上体现了对多发电方和少发电方的公平补偿,还给出了相应补偿公式和阻塞费用计算公式,并证明了阻塞费用等于方案调整后与方案调整前支付费用之差。问题3:采用两种不同方案得到各机组出力分配预案,方案一给出了计算所有段价下各机组能完成的最大负荷的算法,该算法具有一般性,计算量小,并得到负荷需求为982.4MW时清算价是303元/MWh,购电费用74417元,各机组出力为:x1=150,x2=79,x3=180,x4=99.5,x5=125,x6=140,x7=95,x8=113.9方案二采用目标规划方法建立非线性0-1规划模型,采用lingo方便地得到任意负荷下清算价及各机组出力,计算结果与模型一相同。问题4:检验到问题3的分配预案会引起输电阻塞,考虑约束:线路潮流值不超过限值,我们建立了以阻塞费用最小为目标的单目标规划,得到的最小阻塞费用Z=4614.386元,各机组出力方案为:x1=150.688,x2=88,x3=228,x4=80.059,x5=152,x6=96.673,x7=70,x8=117问题5:对负荷需求1052.8MW,我们采用与问题3同样的方法得到清算价为3567元/MWh,购电费用93699元,各机组出力为:x1=150,x2=81,x3=218.2,x4=99.5.x5=135,x6=150,x7=102.1,x8=117检查到该预案会引起输电阻塞,用问题4的单目标模型发现潮流限值内无法调整方案,因此建立阻塞费用最小和各线路上潮流绝对值超过限值的百分比α最小的双目标规划模型,为降低安全隐患,α取最小值5.16%,得到的最小阻塞费用Z=1828.4元,该方案下各台机组出力为:x1=153,x2=88,x3=228,x4=92.107,x5=152,x6=137.354,x7=85.339。
We present the calculating method of jam-up fee, the algorithm of generating sets' pre-distributing contribution scheme and the model of readjusting the pre-distributing scheme in this paper: Problem 1: According to the 32 groups of experimental data, we use multiple linear regression to found the linear representations, which are the 6 main lines' tidal current values with regard to 8 generating sets' contribution and all the representations pass significance test by using SAS8 software. Problem 2: Both out-order capacity and in-order capacity are calculated according to the difference between liquidation price of pre-distributing scheme and quoted price corresponding to the contribution of re-adjusted scheme. It is proved that jam-up fee equals to the charge's difference between pre-distributing scheme and re-adjusted scheme. Problem 3: There are two methods can be used to obtain the pre-distributing contribution scheme of each generating set. One method gives the algorithm of maximum load each generating set can finish under every interval price. This method has universality and small calculating quantity. Another method is to found nonlinear 0-1 programming model by using goal programming. Problem 4: After testing the pre-distributing scheme in problem 3, we know the pre-distributing scheme will cause jam-up. Considering that the lines' tidal current should not exceed restricted value, we found single goal programming model with the target, minimizing jam-up fee. Problem 5: For the load demand 1052.8MW, we obtain the pre-distributing scheme by using the same method in problem 3. Because this scheme will cause jam-up, we found double objectives programming with one target, minimizing jam-up fee and the other target, minimizing α which is the percentage of excessive part, tidal current absolute value exceeding restricted value, to the given restricted value. In order to reduce potential safety hazard, we choose the minimum α=5.16% to get our solution.
出处
《工程数学学报》
CSCD
北大核心
2004年第B12期101-108,共8页
Chinese Journal of Engineering Mathematics
关键词
清算价
序内容量
序外容量
阻塞费用
多元线性回归
目标规划
liquidation price
out-order capacity
in-order capacity
jam-up fee
multiple linear regression
goal programming
