梯级AGC的实现方法研究
|
摘要
由于水电厂固有的运行灵活性和梯级水电厂群之间的补偿协调作用,使梯级水电厂在电力系统的运行中发挥着越来越重要的作用。梯级AGC即为梯级水电厂群的自动发电控制(Automatic Generation Control)。梯级AGC研究的目的主要是降低发电成本,提高发电效益。电力系统中,经济调度的效益相对值虽然不大,但绝对值很大,因此科学合理的进行梯级水电站优化调度具有十分重要的理论意义和广阔的应用前景。国内外学者对梯级AGC问题的研究已经进行了很长时间,也有很多目前仍然没有一套理论上严谨而又很实用的数学模型和算法,很多关于梯级优化的研究一直停留在理论研究阶段。
论文的主要研究成果如下:
回顾了近年来国内外学者在梯级AGC领域取得的新进展,阐述了梯级AGC的关键和存在的不足,指出了梯级AGC研究的发展趋势。
从技术层面上来讲,梯级AGC问题可以分为厂内AGC和厂间AGC两个部分。
针对水电厂发电量一定的前提,建立了厂内AGC的水电厂耗水量最小模型,并采用动态规划法来实现优化调度,对机组启停提出了自己的方法,即在一定的时间内水电厂负荷均满足要求方可停机,并对机组组合、机组负荷安排的优化都进行了考虑。用此方法对清江隔河岩水电厂的历史负荷做方案评价,即用电厂的历史负荷来对安排机组组合和出力,通过比较各台机组历史的和经过优化之后的状态和出力,可知该方法可以创造相当大的经济效益,具有很强的实用性。
根据我国电力系统的实际情况,在各种厂间AGC准则中选择了总蓄能最大准则,并建立了相应的模型,在对准则和模型进行可行有效的简化以后,通过分析各约束条件和比较各种计算方法,选择了一种行之有效的算法——旋转方向法。该方法具有构造方向不需要占用内存的优点,计算简单快速。
最后,对本文的成果进行总结,并提出了有待进一步解决的问题。
Because of the flexibility of the hydraulic power plants, and the compensation and correspondence among cascade hydraulic power plants, the cascade hydraulic power plants are playing a more and more important role in the operation of the power system. The main aim of the research works on AGC (Automatic Generation Control) is to reduce the cost and improve the proficiency. In the power system, the benefit of economical dispatching is not so significant respectively, but would be a big figure absolutely. Thus, the optimized dispatching among cascade hydraulic power plants is very important theoretically and has promising application. Research works on AGC has been carried out for a long period of time, but till now, there is no mathematical model or algorithms that are both practical and precise. The research works remains on the theoretical stage.
The main content of this thesis is as follows:
Some new information on cascade AGC area is reviewed, some key elements and deficiency of AGC are pointed out, and the future trend of AGC is outlined.
Technically, the AGC can be divided into two parts: AGC in and out of hydraulic power plants.
A hydraulic power plant minimum water consumption model of plant AGC is set up, with the power generated by the plants remained constant. Dynamic programming is used in optimizing dispatching, and a new starting-stopping principle is introduced: In a certain period of time, the generator can not be stopped unless all plants have satisfactory loads. The combination of generators and loads dispatching between generators are considered and optimized. This method is used in scheme evaluation for the former loads of Qingjiang- Geheyan hydropower plants, i.e. the former loads of the plants is analyzed with this method to optimize dispatching generator combination and contribution, and then the optimized status and contribution are compared with those real value. The result of this contribution indicates that this method can lead to considerable benefit and has important value in practical operation.
According to the situations in our native power system, the maximum storage rule is selected among various AGC rules. Models are set up accordingly and after simplification on rule and models, analysis on constraint conditions, and comparisons among various algorithms, an effective method-Rosenbrock method is chosen. This method has no memory requirements in calculation, and thus, is easier and faster comparatively.
At last, a conclusion is drawn and some unsettled problems are mentioned.
论文的主要研究成果如下:
回顾了近年来国内外学者在梯级AGC领域取得的新进展,阐述了梯级AGC的关键和存在的不足,指出了梯级AGC研究的发展趋势。
从技术层面上来讲,梯级AGC问题可以分为厂内AGC和厂间AGC两个部分。
针对水电厂发电量一定的前提,建立了厂内AGC的水电厂耗水量最小模型,并采用动态规划法来实现优化调度,对机组启停提出了自己的方法,即在一定的时间内水电厂负荷均满足要求方可停机,并对机组组合、机组负荷安排的优化都进行了考虑。用此方法对清江隔河岩水电厂的历史负荷做方案评价,即用电厂的历史负荷来对安排机组组合和出力,通过比较各台机组历史的和经过优化之后的状态和出力,可知该方法可以创造相当大的经济效益,具有很强的实用性。
根据我国电力系统的实际情况,在各种厂间AGC准则中选择了总蓄能最大准则,并建立了相应的模型,在对准则和模型进行可行有效的简化以后,通过分析各约束条件和比较各种计算方法,选择了一种行之有效的算法——旋转方向法。该方法具有构造方向不需要占用内存的优点,计算简单快速。
最后,对本文的成果进行总结,并提出了有待进一步解决的问题。
Because of the flexibility of the hydraulic power plants, and the compensation and correspondence among cascade hydraulic power plants, the cascade hydraulic power plants are playing a more and more important role in the operation of the power system. The main aim of the research works on AGC (Automatic Generation Control) is to reduce the cost and improve the proficiency. In the power system, the benefit of economical dispatching is not so significant respectively, but would be a big figure absolutely. Thus, the optimized dispatching among cascade hydraulic power plants is very important theoretically and has promising application. Research works on AGC has been carried out for a long period of time, but till now, there is no mathematical model or algorithms that are both practical and precise. The research works remains on the theoretical stage.
The main content of this thesis is as follows:
Some new information on cascade AGC area is reviewed, some key elements and deficiency of AGC are pointed out, and the future trend of AGC is outlined.
Technically, the AGC can be divided into two parts: AGC in and out of hydraulic power plants.
A hydraulic power plant minimum water consumption model of plant AGC is set up, with the power generated by the plants remained constant. Dynamic programming is used in optimizing dispatching, and a new starting-stopping principle is introduced: In a certain period of time, the generator can not be stopped unless all plants have satisfactory loads. The combination of generators and loads dispatching between generators are considered and optimized. This method is used in scheme evaluation for the former loads of Qingjiang- Geheyan hydropower plants, i.e. the former loads of the plants is analyzed with this method to optimize dispatching generator combination and contribution, and then the optimized status and contribution are compared with those real value. The result of this contribution indicates that this method can lead to considerable benefit and has important value in practical operation.
According to the situations in our native power system, the maximum storage rule is selected among various AGC rules. Models are set up accordingly and after simplification on rule and models, analysis on constraint conditions, and comparisons among various algorithms, an effective method-Rosenbrock method is chosen. This method has no memory requirements in calculation, and thus, is easier and faster comparatively.
At last, a conclusion is drawn and some unsettled problems are mentioned.
引文
[1] 邓晓娟,李承军,权先璋等.三峡梯级水电站自动发电控制AGC研究.湖北电力,2003,27(1):4～6
[2] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.142
[3] 罗予如.梯级水电厂群短期经济运行的探讨.水力发电,2000年第2期:57～58
[4] 徐武祥,黄家业,张建文.梯调自动发电控制的实践.云南电业,2002年第七期,46~47
[5] Naccarino J, Cheung R, Briggs W, et al.Real-Time Monitoring, Optimization and Control of a Hydroelectric Complex.IEEE Transactions on Power Systems, 1988,3(4): 1769~1783
[6] 李晖.委内瑞拉EDELCA电力公司梯级与三峡-葛洲坝梯级AGC比较.华中电力,2003(1):31~34
[7] Papageorgiou M.Optimal Multireservoir Network Control by the Discrete Maximum Principle.Water Resources Research, 1985,21(2): 1824~1830
[8] Z.K. Shawwash, T.K. Siu, S.O. Denis.The B.C. Hydro short-term hydro scheduling optimization model.IEEE Transactions on Power Systems, 2000,15(3): 1125-1131
[9] PiekutowskiMR, LitwinowiczT, Frowd R J.Optimal Short-Term Scheduling for a Large-Scale Cascaded Hydro System.IEEE Transactions on Power Systems, 1994,9(2): 805~811
[10] H. Wei, H. Sasaki, J. Kublkawa.Large scale hydrothermal optimal power flow problems based on interior point nonlinear programming.IEEE Transactions on Power Systems, 2000,15(1): 396~403
[11] Habibollahzadeh H, Bubenko J A. Application of Decomposition Techniques to Short Term Operation Planning of Hydro-Thermal Power System. IEEE Transactions on Power Systems, 1986,1(1): 41~47
[12] Xia Q, Xiang N, Wang S, et al.Optimal Daily Scheduling of Cascaded Plants Using a New Algorithm of Non-Linear Minimum Cost Network Flow Concept.IEEE Transactions on Power Systems, 1988,3(3): 929~935
[13] 李伟,纪昌明.梯级水电站群短期经济运行研究——关联预估与微增率逐次逼近相结合法.水电能源科学,2000,18(1):13~15
[14] 朱敏,王孟毅,王定一.梯级水电站日经济运行研究.水电能源科学,1997,15(1):22~28
[15] 梅亚东,朱教新.梯级水电站短期优化调度软件开发及应用介绍.人民珠江,2000,(1):42~44
[16] R. Naresh, J. Sharma.Two-phase neural network based on solution technique for short-term hydrothermal scheduling.Electric Power & Energy Systems, 2002,24: 583~590
[17] 伍永刚,王定一.基于ANN的梯级水电站实时优化运行.系统工程,2000,18(3):78~80
[18] 朱敏,王定一.基于人工神经网络的梯级水电厂日优化运行.电力系统自动化,1999,23(10):35 ~40
[19] M.S. Salam, K.M. Nor, A.R. Hamdam.Hydrothermal Scheduling Based Lagrangian Relaxation Approach to Hydrothermal Coordination.IEEE Transactions on Power Systems, 1998,13(3): 226~234
[20] N.J. Redondo, A.J. Conejo.Short-term hydrothermal coordination by langrangian relaxation: solution of the dual problem.IEEE Transactions on Power Systems, 1999,14(1): 89~95
[21] Y-G Wu, C-Y Ho, D-Y Wang.A diploid genetic approach to short term scheduling of hydrothermal system.IEEE Transactions on Power Systems, 2000,15(4): 1268~1274
[22] 王黎,马光文.基于遗传算法的水电站优化调度新方法.系统工程理论与实践,1997,(7):65~82
[23] 赖一飞,裴金勇,章峰等.基于遗传算法的机组优化组合的应用研究.中国
管理科学,2000,8(2):28~32
[24] 袁晓辉.混沌进划算法及其应用[博士学位论文] .华中科技大学图书馆.2002年
[25] 许还平,朱奕,张彤等.变尺度混沌优化方法在电站经济运行中的应用.哈尔滨工业大学学报,2000,32(4):52~57
[26] 李钰心.水电站经济运行.第一版.北京:中国电力出版社,1999 . 30~34
[27] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.130~135
[28] 姚齐国,张士军,蒋传文.动态规划法在水电站厂内经济运行中的应用.水电能源科学,1999,17(1):46~49
[29] 朱敏.水火电系统中梯级水电厂短期(实时)优化运行控制研究[博士学位论文] .华中科技大学图书馆,1997年
[30] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.77
[31] 陈皓勇,王锡凡.机组组合问题的优化方法综述.电力系统自动化,1999,23(4):51~56
[32] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.126~130
[33] 石琦,李承军,王金文.遗传算法在电力系统日有功优化调度中的应用.电力系统及其自动化学报,2002,14(2):56~59
[34] 左浩,陈昆薇.机组负荷最优分配的改进遗传算法.电力系统及其自动化学报,2001,13(2):16~19
[35] Ross Baldick.The Generalized Unit Commitment Problem.IEEE Transactions on Power Systems, 1995,10(1): 465~475
[36] Chuan-Ping Cheng, Chih-Wen Liu.Unit Commitment by Lagrangian Relaxation and Genetic Algorithms.IEEE Transactions on Power Systems, 2000,15(2): 707 ~714
[37] 王德意,南海鹏.快速动态规划法在运行机组间有功负荷分配中的应用.水利水电技术,2001,32(6):28 ~30
[38] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.142
[39] Walter J.Hobs, Gary Hermon, Stephen Warner.An Enhanced Dynamic programming approach for unit commitment.IEEE Transactions on Power Systems, 1998,3(3): 1201 ~1205
[40] 费修渔.AGC运行对水电机组影响的分析.浙江电力,2001,(4):16 ~18
[41] 张高峰,权先璋,陈建国.三峡梯级水电站自动发电控制.水电自动化与大坝检测,2002,26(2):1~4
[42] H. Habibollahzadeh, D. Frances, U. Sui.A new generation scheduling program at Ontario Hydro.IEEE Transactions on Power Systems, 1990,5(1): 65~73
[43] 徐鼎甲.进行日调节时水电站下游水位的计算.水利水电技术,1995,(4):2 ~4
[44] Naccarino J, Cheung R, Briggs W, et al.Real-time Monitoring, Optimization and Control of a Hydroelectric Complex.IEEE Transactions on Power Systems, 1988,3(4):1769~1783
[45] 朱敏,王定一.电力系统中梯级水电站的日优化运行研究.华中理工大学学报,1997,25(1):47~50
[46] Chow V.T., Heidari, Meredith D.D.Discrete differential dynamic programming approach to water resource systems optimization.Water Resource Research, 1971,7(2):273~282
[47] Mejia M.M.Evaluating Multi-Reservoir Operating Rules.Water Resource Research, 1974,10(6): 125~132
[48] 董子敖.水库群调度与规划的优化理论和应用.济南:山东科学技术出版社, 1989
[49] 粟塔山.最优化计算原理与算法程序设计.第一版.长沙:国防科技大学出版社,2001.87~90
[50] 汪定国.创新与发展——清江体制介绍及入世后的思考.水电能源科学,2002(1):1～5
[2] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.142
[3] 罗予如.梯级水电厂群短期经济运行的探讨.水力发电,2000年第2期:57～58
[4] 徐武祥,黄家业,张建文.梯调自动发电控制的实践.云南电业,2002年第七期,46~47
[5] Naccarino J, Cheung R, Briggs W, et al.Real-Time Monitoring, Optimization and Control of a Hydroelectric Complex.IEEE Transactions on Power Systems, 1988,3(4): 1769~1783
[6] 李晖.委内瑞拉EDELCA电力公司梯级与三峡-葛洲坝梯级AGC比较.华中电力,2003(1):31~34
[7] Papageorgiou M.Optimal Multireservoir Network Control by the Discrete Maximum Principle.Water Resources Research, 1985,21(2): 1824~1830
[8] Z.K. Shawwash, T.K. Siu, S.O. Denis.The B.C. Hydro short-term hydro scheduling optimization model.IEEE Transactions on Power Systems, 2000,15(3): 1125-1131
[9] PiekutowskiMR, LitwinowiczT, Frowd R J.Optimal Short-Term Scheduling for a Large-Scale Cascaded Hydro System.IEEE Transactions on Power Systems, 1994,9(2): 805~811
[10] H. Wei, H. Sasaki, J. Kublkawa.Large scale hydrothermal optimal power flow problems based on interior point nonlinear programming.IEEE Transactions on Power Systems, 2000,15(1): 396~403
[11] Habibollahzadeh H, Bubenko J A. Application of Decomposition Techniques to Short Term Operation Planning of Hydro-Thermal Power System. IEEE Transactions on Power Systems, 1986,1(1): 41~47
[12] Xia Q, Xiang N, Wang S, et al.Optimal Daily Scheduling of Cascaded Plants Using a New Algorithm of Non-Linear Minimum Cost Network Flow Concept.IEEE Transactions on Power Systems, 1988,3(3): 929~935
[13] 李伟,纪昌明.梯级水电站群短期经济运行研究——关联预估与微增率逐次逼近相结合法.水电能源科学,2000,18(1):13~15
[14] 朱敏,王孟毅,王定一.梯级水电站日经济运行研究.水电能源科学,1997,15(1):22~28
[15] 梅亚东,朱教新.梯级水电站短期优化调度软件开发及应用介绍.人民珠江,2000,(1):42~44
[16] R. Naresh, J. Sharma.Two-phase neural network based on solution technique for short-term hydrothermal scheduling.Electric Power & Energy Systems, 2002,24: 583~590
[17] 伍永刚,王定一.基于ANN的梯级水电站实时优化运行.系统工程,2000,18(3):78~80
[18] 朱敏,王定一.基于人工神经网络的梯级水电厂日优化运行.电力系统自动化,1999,23(10):35 ~40
[19] M.S. Salam, K.M. Nor, A.R. Hamdam.Hydrothermal Scheduling Based Lagrangian Relaxation Approach to Hydrothermal Coordination.IEEE Transactions on Power Systems, 1998,13(3): 226~234
[20] N.J. Redondo, A.J. Conejo.Short-term hydrothermal coordination by langrangian relaxation: solution of the dual problem.IEEE Transactions on Power Systems, 1999,14(1): 89~95
[21] Y-G Wu, C-Y Ho, D-Y Wang.A diploid genetic approach to short term scheduling of hydrothermal system.IEEE Transactions on Power Systems, 2000,15(4): 1268~1274
[22] 王黎,马光文.基于遗传算法的水电站优化调度新方法.系统工程理论与实践,1997,(7):65~82
[23] 赖一飞,裴金勇,章峰等.基于遗传算法的机组优化组合的应用研究.中国
管理科学,2000,8(2):28~32
[24] 袁晓辉.混沌进划算法及其应用[博士学位论文] .华中科技大学图书馆.2002年
[25] 许还平,朱奕,张彤等.变尺度混沌优化方法在电站经济运行中的应用.哈尔滨工业大学学报,2000,32(4):52~57
[26] 李钰心.水电站经济运行.第一版.北京:中国电力出版社,1999 . 30~34
[27] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.130~135
[28] 姚齐国,张士军,蒋传文.动态规划法在水电站厂内经济运行中的应用.水电能源科学,1999,17(1):46~49
[29] 朱敏.水火电系统中梯级水电厂短期(实时)优化运行控制研究[博士学位论文] .华中科技大学图书馆,1997年
[30] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.77
[31] 陈皓勇,王锡凡.机组组合问题的优化方法综述.电力系统自动化,1999,23(4):51~56
[32] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.126~130
[33] 石琦,李承军,王金文.遗传算法在电力系统日有功优化调度中的应用.电力系统及其自动化学报,2002,14(2):56~59
[34] 左浩,陈昆薇.机组负荷最优分配的改进遗传算法.电力系统及其自动化学报,2001,13(2):16~19
[35] Ross Baldick.The Generalized Unit Commitment Problem.IEEE Transactions on Power Systems, 1995,10(1): 465~475
[36] Chuan-Ping Cheng, Chih-Wen Liu.Unit Commitment by Lagrangian Relaxation and Genetic Algorithms.IEEE Transactions on Power Systems, 2000,15(2): 707 ~714
[37] 王德意,南海鹏.快速动态规划法在运行机组间有功负荷分配中的应用.水利水电技术,2001,32(6):28 ~30
[38] 张勇传.水电站经济运行原理.第二版.北京:中国水利水电出版社,1998.142
[39] Walter J.Hobs, Gary Hermon, Stephen Warner.An Enhanced Dynamic programming approach for unit commitment.IEEE Transactions on Power Systems, 1998,3(3): 1201 ~1205
[40] 费修渔.AGC运行对水电机组影响的分析.浙江电力,2001,(4):16 ~18
[41] 张高峰,权先璋,陈建国.三峡梯级水电站自动发电控制.水电自动化与大坝检测,2002,26(2):1~4
[42] H. Habibollahzadeh, D. Frances, U. Sui.A new generation scheduling program at Ontario Hydro.IEEE Transactions on Power Systems, 1990,5(1): 65~73
[43] 徐鼎甲.进行日调节时水电站下游水位的计算.水利水电技术,1995,(4):2 ~4
[44] Naccarino J, Cheung R, Briggs W, et al.Real-time Monitoring, Optimization and Control of a Hydroelectric Complex.IEEE Transactions on Power Systems, 1988,3(4):1769~1783
[45] 朱敏,王定一.电力系统中梯级水电站的日优化运行研究.华中理工大学学报,1997,25(1):47~50
[46] Chow V.T., Heidari, Meredith D.D.Discrete differential dynamic programming approach to water resource systems optimization.Water Resource Research, 1971,7(2):273~282
[47] Mejia M.M.Evaluating Multi-Reservoir Operating Rules.Water Resource Research, 1974,10(6): 125~132
[48] 董子敖.水库群调度与规划的优化理论和应用.济南:山东科学技术出版社, 1989
[49] 粟塔山.最优化计算原理与算法程序设计.第一版.长沙:国防科技大学出版社,2001.87~90
[50] 汪定国.创新与发展——清江体制介绍及入世后的思考.水电能源科学,2002(1):1～5