摘要
变电站规划优化是城市配电网规划的重要组成部分,其目的是确定未来电网建设中变电站的站址、站容以及供电范围,避免变电站建设的盲目性,满足未来负荷的发展。
通过研究变电站规划的影响因素、模型分类,确定研究内容及数学模型,并采用拟合规划地区实际建设费用的方法求解目标函数中的费用参数。
针对K-均值聚类算法初值依赖性大的特点,提出一种基于初值均匀冗余分布及其动态减少的初始聚类中心新思想,并把该思想运用到变电站规划中。根据变电站布点的特点,提出一种自动考虑了负荷任意分布情况的冗余网格动态减少的方法求解变电站初始站址。同时,由于模型的简化以及存在最小候选变电站容量,在负荷分布不均匀的情况下可能出现变电站轻负载率而导致变电站不经济运行的现象,引入距离权重因子,有效的平衡了各变电站负载率。另外,在整个变电站规划寻优过程中,引入了多种改进机制,如拐点局部优化机制、变步长寻优机制、供电范围分区机制等,有效的改善了算法的速度和质量。
采用准动态规划法实现多年变电站规划优化。考虑到电网规模的大小对寻优速度的影响,基于目标水平年变电站规划结果,采用两种不同的策略进行中间年变电站规划;当电网规模较小时,采用考虑变电站最优个数的穷举法;当电网规模较大时,考虑寻优速度,提出一种基于负荷适应度的逐年回推法进行解决。
将提出的变电站规划优化算法应用于有275个小区和3761个小区的实际算例中,表明提出的方法在计算精度和速度上都是很有效的。
Substation planning optimization is an important part of urban distribution network planning. It’s purpose is to determine the size, location and power supply range of potential substations in order to avoid the substation construction blindness and satisfy the future load growth.
The research contents and models are determined through the full study of the influence factors and model classification. Various cost coefficients of the objective function are solved by a fitting approach.
Considering the initial value sensitivity of k-means clustering algorithm,a new idea is proposed, which can find a good initial clustering value based on the uniform distribution and redundancy of initial values, and it is applied to the substation planning. According to the characteristics of substation allocation, the Decreasing Redundant Mesh algorithm is proposed, which can automatically consider the load distribution to search for initial substation sites. A method for Alternative Location-Allocation (ALA) is employed to find the size and location of a substation. Meanwhile, an approach of distance weights is employed to balance the loading of individual substations in condition of uneven load distribution. In addition, in the whole process of substation planning, various techniques are used effectively to improve the solution speed and quality, such as the inflection point search, changeable step mechanism and limited supply range.
A Quasi-dynamic Planning method is used to solve the issue of multi-stage substation planning. According to the size of a power grid, two different strategies are adopted. When a power grid is small, an exhaustive method is used considering the optimal number of substations. When a power grid is large, a method of extrapolation is employed considering the load fitness.
The proposed algorithm of the substation planning optimization is applied to the actual examples with 275 loads and 3761 loads. The results show that the proposed method is very effective in calculating precision and speed for solving substation planning optimization problem.
引文
[1]陈章潮,程浩忠.城市电网规划与改造[M].北京:中国电力出版社,2007.
[2] E.Lakervi,M.Nurmi. Present status in applying mathematical planning methods in AM/FM/GIS system[C].International Conference on Electricity Distribution-CIRED Birtningham(UK),1997:2-5.
[3]城市电力网规划设计导则[S].原能源部、建设部联合颁发的能源电(1993)228号文.
[4]高炜欣,罗先觉,朱颖.贪心算法结合Hopfield神经网络优化配电变电站规划[J].电网技术,2004,28(7):73-76.
[5]周敏,程浩中,周中明.优化理论在城市配电网变电站选址中的应用[J].电力科学与工程,2003,2(2):36-37.
[6] E.Lakervi,E.J.Holmes著,范明天,张祖平,岳宗斌译.配电网络规划与设计[M].北京:中国电力出版社,1999.
[7] MASUD E. An Interactive Procedure for Sizing, and Timing Distribution Substations Using Optimal Techniques [J]. IEEE Trans on Power Apparatus and Systems, 1974, 93(5): 1281-1286.
[8]张焰,张国慧.电网规划中的最小模糊缺电成本计算研究[J].系统工程理论与实践, 2000,35(7):123-127.
[9] Levi V A,Calovic M S. A new decomposition based method for optimal expansion planning of large transmission network[J]. IEEE Transaction on Power Systems, 1993,6(3):937-943.
[10] H.K.Temraz,M.M.A.Salama. A Planning Model for Siting, Sizing and Timing of Distribution Substation and Defining the Associated Service Area [J]. Elsevier Power System Research,2002,62: 145-152.
[11]刘巍,章健.变电站站址优化方法的研究[J].郑州轻工业学院学报(自然科学版),2003,18(1):31-34.
[12]邓先礼等.最优化技术[M].重庆:重庆大学出版社,1998.
[13] Lee T Y ,Hick K L. Transmission Expansion by Branch-Bound Integer Programming with Optimal Cost*Capacity Curves[J]. IEEE Transaction on Power Systems , 1974, 93( 5):123-128.
[14] E.Masud. An Interactive Procedure for Sizing and Timing Distribution Substation Using Optimization Techniques[J]. IEEE Trans,PAS 93,1974:1281-1286.
[15] T.Gonen, I.L.Ramirez-Rosado. Optimal Multi-stage Planning of Power Distribution System[J]. IEEE Transaction on Power Delivery 2(2),1987:512-519.
[16] Thompson G L,Wall D L. A Branch and Bound Model for Choosing Optimal Substation Locations[J]. IEEE Trans on Power Apparatus and Systems, 1981,100(5) : 2683_2688.
[17] Laura B, Gerson C O, Mario P. A mixed integer disjunctive model for transmission network expansion[J]. IEEE Transactionson Power Systems, 2001, 16(3):560-565.
[18] Hindi K S, Brameller. Design of low-voltage distribution network: a mathematical programming method[J]. IEE.Proc-C,1977,124(1):54-58.
[19]王家耀.空间信息系统原理[M].北京:科学出版社,2001.
[20] Held.M, Williamson.R.B. Creating electrical distribution boundaries using computational geometry, Power Systems[J]. IEEE Transactions,2004,19(3):1342– 1347.
[21]黄小铼,陈星莺.基于负荷转移的中压配电所综合规划[J].电力自动化设备,2005,25(4):62-64.
[22]杨丽徙,朱向前,陈根永,王金凤.基于Vorono i图的变电站优化规划方法[J].电力系统及其自动化学报,2006,18(5):10-13,20.
[23]葛少云,李慧,刘洪.基于加权Voronoi图的变电站优化规划[J].电力系统自动化,2007,31(3):29-34.
[24] L.Cooper. Location-allocation problem[J].Operation Research,1963,11(3):331-343.
[25]玄光男(日),程润伟著,汪定伟,唐加福,黄敏译译.遗传算法与工程设计[M].北京:科学出版社,2000,257-261.
[26]汤红卫,王华,郭喜庆.一种基于地理信息系统的配电网规划方法[J].电网技术,2002,26(12):79-82.
[27] Lo. K.L,Nashid. I. Interactive expert system for optimal design of electricity distribution systems[J]. IEE Proceedings,1996 ,143( 2):151-156.
[28]林敏.配电网优化规划中变电站站址及站间连接的选择[J].江苏电机工程,2001,20(3):6-9.
[29] D.Hongwei, Y.Yixin, H.Chunhua, et al. Optimal Planning of Distribution Substation Locations and Sizes-model and Algorithm[J]. Electrical Power & Energy Systems, 1996,18(6): 353-357.
[30]余贻鑫,张崇见.城市电网中压配电变压器的模糊优化规划[J].中国电机工程学报,2000,20(4):23-28.
[31]张崇见,余贻鑫,严雪飞.中压配电变压器优化规划[J].电力系统自动化,1999,23(1):27-30.
[32]董永峰,杨彦卿等.基于改进粒子群算法的变电站选址规划[J].继电器,200 8,36(5):32-35.
[33]严司玮,姚建刚等.基于改进粒子群算法的变电站两阶段优化选址[J].电力系统保护与控制,2010,38(5):34-38.
[34]徐珍霞顾洁.离散粒子群优化算法在变电站选址中的应用[J],电气应用,2006,25(4):35-38,115.
[35]路志英,葛少云,王成山.基于粒子群优化的加权伏罗诺伊图变电站规划[J].中国电机工程学报2009,29(16):35-41.
[36] Wang Zhiqiang, Que Huakun, Zhang Lixin。Substation plan based on improved model and Adaptive Mutation Particle Swarm Optimization Algorithm [J].Intelligent Systems and Applications, 2009, International Workshop on:1-4.
[37] Rainer Storn, Kenneth Price . Differential evolution- a simple and efficient heuristic for global optimization over continuous spaces [J]. Journal of Global Optimization , 1997, 11( 4) : 341- 359 .
[38]周清清。微分进化算法在变电站选址中的应用[J],电气开关,2009,2:56-59
[39]牛卫平,刘自发等,基于GIS和微分进化算法的变电站选址及定容[J],电力系统自动化,2007,31(18):82-86.
[40]许童羽,孙艳辉,基于GIS和改进微分进化算法的农网变电站选址定容[J].电力系统保护与控制,2009,37(22):34-39
[41] Nian Liu, Jianhua Zhang, Wenxia Liu。A Spatial Information-based Self-Adaptive Differential Evolution for Distribution Substations Location and Sizing ,2008 IEEE Pacific-Asia Workshop on Computational Intelligence and Industrial Application.
[42]陈根军,王磊,唐国庆.基于蚁群最优的输电网络扩展规划[J].电网技术,2001,25(6):21-24.
[43]王志刚,杨丽徙,陈根永.基于蚁群算法的配电网网架优化规划方法[J].电力系统及其自动化学报,2002,14(6):73-76.
[44]陈根军,唐国庆.基于禁忌搜索与蚁群最优结合算法的配电网规划[J].电网技术,2005,29(2):23-27.
[45] Peter M, Bernd F. A comparison of memetic algori-thms, Tabu search, and antcolonies for the quadratic assignment problem [J]. Proceedings of the 1999 Congress onEvolutionary Computation Volume 3,1999,1(1):2063-2070.
[46] Luk S, Bart N. Ant colonies are good at solving constraint satisfaction problems [A].In: Proceedings of the 2000 Congression Evolutionary Computation Volume 2 [C]. SanDiego, USA: 2000.1190-1195.
[47]谢敬东,王磊,唐国庆.遗传算法在多目标电网规划中的应用[J].电力系统自动化.1998,22(10):20-22.
[48]鞠平,王勇,李靖霞.配电网综合规划的遗传算法[J].河海大学学报(自然科学版),2003,31(1):87-90.
[49]刘可真,陈勇,束洪春,基于改进遗传算法的输电网优化规划[J],昆明理工大学学报(理工版),2007,32(1):31-35
[50]程浩忠,高赐威,马则良.多目标电网规划的一般最优化模型[J].上海交通大学学报,2004,38(8):1229-1233.
[51]顾洁,陈章潮等.混合遗传-模拟退火算法在电网规划中的应用[J].上海交通大学学报, 1999,33(4):485-487.
[52]邢文训,谢金星.现代优化计算方法[M].北京:清华大学出版社,1998.
[53] R Ramero, R A Gallego. A Monticelli. Transmission System Expansion Planningby Simulated Annealing[J]. IEEE Transactions on Power Systems, 1996,1(11):222-228.
[54]金明.基于模拟退火算法的输电网中长期规划研究[J].青岛大学学报.1999,12(1):80-85.
[55]陈章潮,顾洁,孙纯军.改进的混合模拟退火-遗传算法应用于电网规划[J].电力系统自动化. 1999,23(10):28-32.
[56]陈根军,李继光,唐国庆.基于Tabu搜索的配电网络重构算法[J].中国电机工程学报, 2002,22(10):28-33.
[57]陈根军,李继光,王磊等.基于Tabu搜索的配电网络规划[J].电力系统自动化, 2001,25(7)40-44.
[58]高炜欣,罗先觉.基于Hopfield神经网络的多阶段配电变电站的规划优化[J].电工技术学报,2005,20(5):58-64
[59]高炜欣,罗先觉,朱颖.贪心算法结合Hopfield神经网络优化配电变电站规划[J].电网技术,2004,28(7):73-76.
[60]方兴,郭志忠.配电网规划研究述评[J].电力自动化设备,2003,23(5):71-74
[61]孔涛,程浩忠,李钢,谢欢.配电网规划研究综述[J].电网技术,2009,33(19):92-99.
[62]牛辉,程浩忠,张焰等.电网扩展规划的可靠性和经济性研究综述[J].电力系统自动化,2000,24(1):51-56.
[63] J Han J,Kamber M.范明,孟小峰,等译.数据挖掘概念与技术[M].北京:机械工业出版社,2006.
[64]李东琦.聚类算法的研究[D],西南大学硕士学位论文,2007.
[65] Eilon, Samuel.Distribution Management:Mathematical Modeling and Practical Analysis[M]. London: St. Martin's Griffin,1971.
[66]谢云等.非数值并行算法(第一册)[M].科学出版社,1994.
[67]蒲繁强.基于分区加权Voronoi图的变电站选址研究[D],天津大学,2008