摘要
在开放的电力市场,电力配电系统起着非常重要和关键的作用。配电公司在电力市场中为用户提供输配电服务,是联系用户和电力系统的桥梁和纽带。它能影响市场的效率,并使市场更加可靠、安全和有益于各市场成员。因此,在电力市场环境下需对配电公司的运营作详细的分析和研究,以改进其短期和长期运营策略。
论文按负荷预测、配电系统运营、可中断负荷及意义分类综述了已有的研究成果。配电系统具有结构复杂和设备量巨大的特点,导致其分析难度较大。放射式结构的广泛使用和馈线支路参数电阻与电抗比值较高的普遍性,造成其解决方法的局限性。潮流求解的结果可得到电压大小、电压角度、线路功率等参数。最优潮流是从电力系统稳定运营的角度来调整系统中各种控制设备的参数,在满足节点正常功率平衡及各种安全指标的约束下,实现目标函数最小化的优化过程；通常的目标函数是发电费用、发电耗量或全网的网损。论文从经济的角度出发,以配电系统总损耗作为目标函数,用33节点配电网络作为算例,利用配电最优潮流模型进行潮流计算。
为了提高短期负荷预测速度和精度,提出了将模糊聚类分析和神经网络代数算法相结合的短期负荷预测方法。论文综合考虑天气、日类型、历史负荷等对未来负荷变化的影响,通过模糊聚类分析选取学习样本,找出与预测日相符的预测类别,采用神经网络代数算法训练样本,对24小时负荷(24点)每点建立一个预测模型。该方法充分发挥了神经网络和模糊理论处理非线性问题的能力,提高了学习效能,而且克服了传统反向传播算法存在的缺点。算例分析结果表明该方法有较高的预测精度,取得了令人满意的结果。
论文提出了电力零售市场环境下配电系统总体运营框架。该运营框架是一个两阶段层次模型,第一阶段称为配电系统的第一阶段/目前运营模式。第二阶段涉及配电系统实时运营,称为第二阶段/实时运营模式。考虑需求波动或电力市场价格的变化,配电系统的运营架构应包括可供选择的日前24小时的电力供应准备方案,配电系统还必须监控时刻变化的批发市场价格和准确预测用户需求。
日前运营模式的决策结果包括：网络购买、调度自备的分布式电源、订立可中断负荷合同,决策结果作为实时运营模式的约束；配电系统实时操作受到网络配置和需求响应的影响。系统存在分布式电源,并存在与用户签订可中断合同可能性,馈线潮流将明显不同于传统的配电系统。因此,配电系统支路损耗,电压分布和支路容量的余裕都将受到影响。配电系统十分注重日前决策,旨在使其短期成本最小化。论文用33节点配电网络作为算例对两阶段运营决策模型进行了研究,并构建了二个不同运营方案以分析配电公司的行为和决策。
考虑到存在分布式电源及其对配电系统的影响,论文提出了将分布式电源敏感度系数概念纳入配电系统运营模式。敏感度系数反映了分布式电源对配电系统损耗的影响。在配电系统中,分布式电源对有功和无功功率损耗的影响的变化量被称为有功/无功增量损耗指数。考虑了敏感度系数的配电系统运营模式,旨在减少配电系统短期能源成本。论文用18节点配电网络作为算例进行了研究。
包括太阳能、风能和小水电等具有不同技术特点的分布式电源参与配电系统的运营,急需一系列适应电力市场运营的理论及技术,包括竞价交易模式、电力市场技术支持系统等,以促进资源的优化配置、提高效率和降低成本。论文提出了具有不同技术特点的分布式电源参与的基于“容量合同+效率置换+实时市场”的发电竞价模式,力求为分布式电源发电竞价上网打下技术基础,为完善电力市场的管理发挥作用。
In an open electricity market, power distribution system plays a very important and key role, distribution companies in the electricity market to provide users with transmission and distribution services, is to contact the user and a bridge linking the power systems. It can affect the efficiency of markets and to market more reliable, safe and beneficial to all market members. Therefore, in the electricity market environment, distribution companies need to operate on a detailed analysis and research to improve its short-term and long-run strategy.
Paper by load forecasting, distribution system operators, interruptible load, and meanings, and classification review of the existing research results.Trend of the results obtained solving the voltage magnitude, voltage angle, line power and other parameters. Distribution systems and equipment with a huge amount of complex structure characteristics, resulting in their analysis difficult. Structure of the widespread use of radiation and the feeder slip resistance and reactance parameters of the universality of the ratio higher, resulting in limitations of their solutions. Power system optimal power flow is from the perspective of stable operation of various control devices to adjust the system parameters to meet the normal nodes and various safety indicators power balance constraints, to achieve the objective function to minimize the optimization process; the usual objective function power generation costs, power consumption or the entire network of net losses. Papers from the economic point of view, the distribution system as the objective function of total loss, with33nodes distribution network as an example, optimal power flow model using distribution power flow calculation.
In order to improve the speed and accuracy of short-term load forecasting,considering the combined influence of weather, day type and historical load data, a new short-term load forecasting method is put forward in which the learning samples are selected by the fuzzy clustering,finding out the category coincident with that of the daily load to be forecasted and using the neural network algebra algorithm, a forecasting model for each point in24points are established. This method gives sufficient play to the ability of processing non-linear problems by neural network and fuzzy theory, the learning efficiency are improved,and completely overcome the shortcoming of the traditional BP algorithm. The results show that the proposed method possesses better forecasting accuracy and the forecasting is satisfactory.Thesis, the retail electricity market operating under the framework of the overall distribution system. The operating framework is a two-stage hierarchical model, the first stage, known as the distribution system first stage operating mode/day ahead operating mode (FSOM/DAOM). The second stage involves real-time power distribution system operation, known as second stage operating mode/real-time operating mode (SSOM/RTOM). Consider the needs of the electricity market price fluctuations or changes in the operating structure of distribution system should include alternative electricity supply24hours before the program of preparation and distribution system must also monitor the ever-changing wholesale market prices and predict customer needs.
FSOM/DAOM decision results, including:Online Shopping, scheduling their own distributed generating units, interruptible load contracts entered into the decision-making results as SSOM/RTOM constraints; distribution system network configuration and real-time operating under the impact of demand response. There is a distributed system unit power, and there can interrupt the contract signed with the user the possibility of the trend will be significantly different from the traditional feeder distribution system. Therefore, the slip distribution system losses, voltage distribution and branch capacity margin will be affected. Attaches great importance to decision-making power distribution system has to make short-term costs are minimized. Paper distribution network with33nodes as an example of a two-stage operational decision-making model was studied, and construct two different programs to analyze the distribution companies operating behavior and decision-making.
Taking into account the existence of a distributed generator and the impact on the distribution system, the paper will be distributed generators into the distribution system sensitivity coefficient of the concept of operating modes. Sensitivity coefficient reflects the distribution systems of distributed generators on the impact of loss. In the distribution system, distributed generators active and reactive power on the impact of the change in loss is known as active/reactive power incremental loss index. Consider the sensitivity coefficient distribution system operating mode, power distribution systems designed to reduce short-term energy costs. Paper distribution network with18nodes was studied as an example.
Including solar, wind and small hydropower stations with different technical characteristics of the DG units in distribution system operations, need to adapt to the electricity market operated by a series of theory and technology, including the auction mode, the power market technical support system, so as to promote the resources optimal allocation, increase efficiency and reduce costs. Thesis with different technical characteristics of DG units to participate based on "Capacity Contract+Efficiency displacement+Timely market," the power auction model, and strive for the DG unit power generation technology Bidding lay foundation for improving the management of the electricity market to play a role.
引文
[1]姚建刚,章建.电力市场分析[M].北京：高等教育出版社,1999,20-26
[2]姚建刚,章建等.电力市场运营及其软件开发[M].北京：中国电力出版社,2002,192-207
[3]赵豫,于尔铿.电力零售市场研究(二)电力零售市场的组成[J]电力系统自动化,2003,Vol.27 No.10：35-38
[4]李渝曾,王睍,张少华计及输电约束的区域双边电力市场LSFE分析[J]电力系统自动化,2005,Vol.29 No.2：8-12
[5]周淙,李健,孙才新等.基于粗糙集和元胞自动机的配电网空间负荷预测[J],中国电机工程学报,2008,28(25)：68-73.
[6]霍利民,范新桥,黄丽华等.基于基因表达式程序设计及误差循环补偿的电力系统短期负荷预测[J],中国电机工程学报,2008,28(28)：103-107
[7]陶文斌,张粒子,潘弘等.基于双层贝叶斯分类的空间负荷预测[J],中国电机工程学报,2007,27(7)：13-17
[8]康重庆,牟涛,夏清.电力系统多级负荷预测及其协调问题(一)研究框架[J],电力系统自动化,2008,32(7)：34-38
[9]牟涛,康重庆,夏清.电力系统多级负荷预测及其协调问题(二)基本协调模型[J],电力系统自动化,2008,32(8)：14-18
[10]王雪峰,邬建华,冯英浚等.运用样本更新的实时神经网络进行短期电力负荷预测[J]系统工程理论与实践,2003年4月第4期：95-99
[11]李永坚,胡鹤宇.电力系统短期负荷预测的级联网络模型研究[J]继电器(RELAY),2004, Vol.32 No.10:14-18
[12]张海涛,陈宗海,朱六璋.基于改进FLN的短期电力负荷预测算法[J]电工技术学报,2004,Vol.19 No.5：92-96
[13]谢宏,程浩忠,张国立.基于粗糙集理论建立短期电力负荷神经网络预测模型[J]中国电机工程学报,2003,Vol.23 No.11：1-4
[14]I.Orezga, S.Rahman. Input Variable Selection for ANN-Based Short-Term Load Forecasting. IEEE Trans on Power Systems,Vol.13, No.4, pp.1238-1244, November 1998
[15]杨靖研,杜德生.一种电力系统短期负荷预测的新方法[J]控制理论与应用,2004年第23卷第2期：14-17
[16]Bin Ye; Chuangxin Guo; Yijia Cao. Short-term load forecasting using a new fuzzy modeling strategy. Intelligent Control and Automation,2004. WCICA 2004. Fifth World Congress on,Volume6, June15-19,2004 Pages:5045-5049
[17]Yuancheng Li; Bo Li; Tingjian Fang. Short-term load forecast based on fuzzy wavelet support vector machines. Intelligent Control and Automation,2004. WCICA 2004. Fifth World Congress on, Volume6, June 15-19,2004 Pages: 5194-5198
[18]Al-Kandari, A.M.; Soliman, S.A.; E1-Hawary, M.E. Fuzzy systems application to electric short-term load forecasting. II. Computational results. Power Engineering,2003 Large Engineering Systems Conference on,7-9 May 2003 Pages:131-137
[19]姜勇.基于模糊聚类的神经网络短期负荷预测方法[J]电网技术.2003,Vo1.27,No.2：45-49
[20]谢宏,陈志业,牛晓东等基于小波分解与气象因素影响的电力系统日负荷预测模型研究[J] 中国电机工程学报,2001,21(5)：5-10
[21]高山,单渊达小波奇异性检测在负荷数据纠错和平滑处理中的应用[J]中国电机工程学报,2001,21(11)：105-108
[22]邰能灵：侯志俭,李涛等基于小波分析的电力系统短期负荷预测方法[J]中国电机工程学报,2003.23(1)：45-49
[23]黎灿兵,李晓辉,赵瑞等.电力短期负荷预测相似日选取算法[J],电力系统自动化,2008,32(9)：69-73
[24]王鹏,邰能灵,王波等.针对气象因素的短期负荷预测修正方法[J],电力系统自动化,2008,32(13)：92-96
[25]李鹰,赵振江,吴松涛灰色模型在普通日短期电力负荷预测中的应用[J]长沙电力学院学报(自然科学版),2003,18(1)：15-17
[26]符杨,朱兰,曹家麟.基于模糊贴近度理论的负荷密度指标求取新方法[J],电力系统自动化,2007,31(19)：46-49
[27]牛东晓,赵海青.负荷预测的交叉式自适应优选组合预测模型[J]华北电力大学学报,2000 Vo1.27 No.04：13-17
[28]夏道止,王锡凡,周琳等.基于RBF神经网络和专家系统的短期负荷预测方法[J]西安交通大学学报,2001,35(04)：331-334
[29]朱向阳,林鹤云一种基于RAN和专家系统的短期负荷预测方法[J]华东电力,2004,32(9)：42-44
[30]俞集辉,李春燕,谢开贵.基于遗传算法的短期负荷组合预测模型[J]电网技术2001,25(8)：20-23
[31]王孙安,盛万兴,尤勇.一种新型短期负荷预测模型的研究及应用[J]中国电机工程学报,2002,22(09)：15-18
[32]Hiroyuki Mori,Shouichi Urano. Short-Term Load Forevasting with Chaos Time Series Analysis.:133-137,May 2010 Pages:131-137
[33]张智晟,孙雅明,王兆峰,李芳.优化相空间近邻点与递归神经网络融合的短期负荷预测[J]中国电机工程学报,2003,23(8)：44-49
[34]杨薛明,苑津莎,王剑锋等.基于云理论的配电网空间负荷预测方法研究[J]中国电机工程学报,2006,,26(6)：30-36
[35]吕金虎,张锁春.加权一阶局域法在电力系统短期负荷预测中的应用[J] 控制理论与应用,2002,19(5)：767-770
[36]肖鑫鑫,刘东.分布式供能系统接入电网模型研究综述,[J]华东电力,2008,Vol.36 No.2 Feb.76-81
[37]A.Bhattacharya, P.K.Chattopadhyay, Application of biogeography-based optimisation to solve different optimal power flow problems[J]. IET Generation, Transmission & Distribution, Vol.5, No.1,pp:70-80, January.2011.
[38]A.A. Romero,H.C. Zini.G. Ratta,R. Dib. Harmonic load-flow approach based on the possibility theory [J]. IET Generation, Transmission & Distribution, Vol.5, No.4, pp.393-404,April 2011.
[39]Andrea A. Sousa, Geraldo L. Torres, and Claudio A. Canizares, Robust Optimal Power Flow Solution Using Trust Region and Interior-Point Methods[J].IEEE Transactions on Power Systems, Vol.26, No.2, pp.487-499, May 2011.
[40]R. A. Jabr, "Radial distribution load flow using conic programming," IEEE Transactions on Power Systems, Vol.21, pp.1458-1459, August 2006.
[41]Y. Deng, X. Ren, C. Zhao, and D. Zhao, "A heuristic and algorithmic combined approach for reactive power optimization with time-varying load demand in distribution systems", IEEE Transactions on Power Systems, Vol.17, pp.206-1231, Nov.2002.
[42]S. F. Mekhamer, S. A. Soliman, M. A. Moustafa, and M. E. E1-Hawary, 'Application of fuzzy logic for reactive-power compensation of radial distribution feeders", IEEE Transactions on Power Systems, Vol.18, pp. 206-1231, February 2003
[43]A. Ahuja and A. Pahwa, "Using Ant Colony Optimization for Loss Minimization in Distribution Networks," Proceedings of the 37th North American Power Symposium, Iowa State University, pp.470-474, Oct.2005
[44]D. Zhang, Z. Fu, and L. Zhang, "Joint Optimization for Power Loss Reduction in Distribution Systems", IEEE Transactions on Power Systems, vol.23, no.1, pp. 161-169, Feb.2008
[45]S. Mandal and A. Pahwa, "Optimal Selection of Conductors for Distribution Feeders," IEEE Transactions on Power systems, pp.192-197. Feb.2002.
[46]S. Segura, L.C.P. da Silva. R. Romero.Generalised single-equation load flow method for unbalanced distribution systems [J]. IET Generation, Transmission & Distribution, Vol.5. No.3, pp.347-355, March 2011
[47]P. M. De Oliveira-De Jesus, M. T. Ponce de Leao, J. M. Yusta, H. M. Khodr, and A. J. Urdaneta, "Uniform marginal pricing for the remuneration of distribution networks," IEEE Transactions on Power Systems, Vol.20, pp.1302-1310, August 2005.
[48]P. M. Sotkiewicz and J. M. Vignolo, "Nodal pricing for distribution networks: efficient pricing for efficiency enhancing DG," IEEE Transactions on Power Systems, Vol.21, pp.1013-1014. May 2006.
[49]E. Carpaneto, G. Chicco, and J. S. Akilimali, "Branch current decomposition method for loss allocation in radial distribution systems with distributed generation," IEEE Transactions on Power Systems, Vol.21, pp.1170-1179, August 2006
[50]P. M. Sotkiewicz and J. M. Vignolo, "Allocation of fixed costs in distribution networks with distributed generation," IEEE Transactions on Power Systems, Vol.21, pp.639-652, May 2006.
[51]S. Tong and K. N. Miu, "A network-based distributed slack bus model for DGs in unbalanced power flow studies," IEEE Transactions on Power Systems, Vol. 20, pp.835-842, May 2005.
[52]R. Palma-Behnke, J. L. Cerda A., L. S. Vargas, and A. Jofre, "A distribution company energy acquisition market model with integration of distributed generation and load curtailment options," IEEE Transactions on Power Systems, Vol.20, pp.1718-1727. November 2005.
[53]P. M. Costa and M. A. Matos, "Loss allocation in distribution networks with embedded generation", IEEE Transactions on Power Systems, Vol.19, pp.384-389, February 2004
[54]W. El-Khattam, Y. Hegazy and M. M. A. Salama, "Investigating Distributed Generation Systems Performance Using Monte Carlo Simulation," IEEE Transactions on Power Systems, vol.21, pp.524-532, May 2006.
[55]N. C. Scott, D. J. Atkinson, and J. E. Morrell, "Use of load control to regulate voltage on distribution networks with embedded generation", IEEE Transactions on Power Systems, Vol.17, pp.510-515, May.2002.
[56]M. Thomson and D. G. Infield, "Network power flow analysis for a high penetration of distributed generation", IEEE Transactions on Power Systems, vol. 22, pp.1157-1162, Aug.2007
[57]H. A. Gil and G. Joos, "On the quantification of the network capacity deferral value of distributed generation," IEEE Transactions on Power Systems, Vol.21, pp.1592-1599, Nov.2006.
[58]C. Wang and M. H. Nehrir, "Analytical approaches for optimal placement of distributed generation sources in power systems", IEEE Transactions on Power Systems, Vol.19, pp.2068-2076, November 2004.
[59]W. El-Khattam, Y. Hegazy and M. M. A. Salama, "An integrated distributed generation optimization model for distribution system planning,"IEEE Transactions on Power Systems, vol.20, pp.1158-1165, May 2005.
[60]W. El-Khattam, K. Bhattacharya, Y. Hegazy and M. M. A. Salama, "Optimal investment planning for distributed generation in a competitive electricity market," IEEE Transactions on Power Systems, vol.19, pp.1674-1684, August 2004.
[61]A. Chowdhury, S. K. Agarwal, and D. O. Koval, "Reliability modeling of distributed generation in conventional distribution systems planning and analysis," IEEE Ind. Appl. Mag., vol.39, no.5, pp.1493-1498, Sep.2003.
[62]S. McCusker and B. F. Hobbs, "A nested decomposition approach to locating distributed generation in a multiarea power system," Networks and Spatial Economics, vol.3, no.2, pp.197-223, Jun.2003.
[63]张涛,宋家骅,程晓磊等,电力市场环境下可中断负荷综述,[J]吉林电力,Vol.35 No.2(Ser.No.189),25-28,Apr.2007
[64]P. Jazayeri, A. Schellenberg, W. D. Rosehart, J. Doudna, S. Widergren, D. Lawrence, J. Mickey, and S. Jones," A Survey of load control programs for price references 160 and system stability," IEEE Transactions on Power Systems, Vol. 20, pp.1504-1509, August 2005.
[65]L. A. Tuan, K. Bhattacharya, "A review on interruptible load management: literature and practice", in Proc. of 33rd North America Power Symposium, Texas, USA, pp.406-413, October 2001.
[66]K. Bhattacharya, M. H. J. Bollen and J. E. Daalder, "Real-time optimal interruptible tariff mechanism incorporating utility-customer interaction," IEEE Transactions on Power Systems, Vol.15, pp.700-706, May 2000.
[67]L. A. Tuan and K. Bhattacharya, "Competitive framework for procurement of interruptible load services." IEEE Transactions on Power Systems, vol.18, pp. 889-897, May 2003.
[68]M. Fahriog-lu, and F. L. Alvarado, "Designing incentive compatible contracts for effective demand management", IEEE Transactions on Power Systems, Vol. 15, pp.1255-1260, Nov.2000.
[69]A. K. David, "Load forecasting under spot pricing." IEE Proceedings, Pt. C, vol. 135, no.5, pp.369-377, Sept.1988.
[70]A. K. David and Y. Z. Li, "Consumer rationality assumptions in the real-time pricing of electricity," IEE Proceedings, Pt. C, vol.139, no.4, pp.315-322, July 1992.
[71]E. Bompard, E. Carpaneto, G. Chicco, and G. Gross, "The role of load demand elasticity in congestion management and pricing", in Proc. of IEEE Power Engineering Society Summer Meeting, Vol.4, pp.2229-2234,2000.
[72]T. Niimura, M. Dhaliwal, K. Ozawa, "Evaluation of retail electricity supply contracts in deregulated environment" IEEE Power Engineering Summer Meeting, Vol.2, pp.1058-1062.2001.
[73]S.-E., Fleten, and E. Pettersen, "Constructing bidding curves for a price-taking retailer in the Norwegian electricity market"', IEEE Transactions on Power System, Vol.20, pp.701-708. May.2005.
[74]Farrokh Aminifar, Mahmud Fotuhi-Firuzabad, Mohammad Shahidehpour. Unit Commitment With Probabilistic Spinning Reserve and Interruptible Load Considerations[J].IEEE Transactions on Power System, Vol.24, No.1, pp.388-397, February 2009
[75]Haiying Li, Yuzeng Li, Zuyi Li. A Multiperiod Energy Acquisition Model for a Distribution Company With Distributed Generation and Interruptible Load[J]. IEEE Transactions on Power Systems, Vol.22, No.2, pp.588-596, May 2007
[76]都亮,刘俊勇,田立峰等.电力市场环境下秒级可中断负荷研究[J],中国电机工程学报,2008,28(16)：90-95
[77]李海英,李渝曾,张少华.具有分布式发电和可中断负荷选择的配电公司能量获取模型[J],中国电机工程学报,2008,28(10)：88-93
[78]薛禹胜,罗运虎,李碧君等.关于可中断负荷参与系统备用的评述[J],电力系统自动化,2007,31(10)：1-6
[79]罗运虎,薛禹胜.Gerard LEDWICH等.低电价与高赔偿2种可中断负荷的协调[J],电力系统自动化,2007,31(11)：17-21
[80]李金波,张少华.考虑用户风险偏好的可中断负荷定价[J],电网技术,2008, 32(3)：52-55
[81]王瑞庆,李渝曾,张少华.考虑分布式发电和可中断负荷的配电公司购电组合策略研究[J],电力系统保护与控制,2009,37(22)：17-21
[82]J. D. Glover and M. S. Sarma, Power System Analysis and Design, Third edition, 2002.
[83]李妮,江岳春,黄珊.基于累积式自回归动平均传递函数模型的短期负荷预测[J]电网技术,2004,28(8)：30-33.
[84]江岳春,姚建刚,毛弋.基于“容量合同+效率置换+实时市场”模式的发电竞价系统的研究[J]电工技术学报2006 21(1)：52-57.
[85]毛李帆,江岳春,姚建刚等.采用正交信号修正法与偏最小二乘回归的中长期负荷预测[J]中国电机工程学报2009 29(16)：82-88.
[86]江岳春,姚建刚,毛弋.电网发电竞价上网数据申报及信息发布系统[J]湖南大学学报200431(3)：43-46.
[87]陈伟,吴耀武,娄素华等.基于累积式自回归动平均法和反向传神经网络的短期负荷预测模型[J]电网技术,2007,31(3)：73-76.
[88]毛李帆江岳春龙瑞华等.基于偏最小二乘回归分析的中长期电力负荷预测[J]电网技术2008 32(19)：71-77
[89]金海峰,熊信艮,吴耀武.基于级联神经网络的短期负荷预测方法[J]电网技术,2002,26(3)：49-51.
[90]马文晓,白晓民,沐连顺.基于人工神经网络和模糊推理的短期负荷预测方法[J]电网技术,2003,27(5)：29-32.
[91]鞠平,姜巍,赵夏阳等.96点短期负荷预测方法及其应用[J]电力系统自动化,2001,25(22)：32-36.
[92]刘亚,张国忠,何飞.节气负荷预测方法研究[J]电力自动化设备,2003,23(7)：39-42.
[93]Shimrhmamdai D,Wayne Hong H. Reconfiguration of Electric Disrtibution Networks for Resistive Line Losses Reduction. IEEE Trnasactions on Power Deliveyr,1989,4(2): 1492-1498
[94]GAMS Release 2.25", in A User's Guide, GAMS Development Corporation, 1998