基于改进引力搜索算法的风电功率短期预测优化调度研究
|
摘要
考虑到风电功率本身的随机波动特性,在制定日前市场经济预调度方案过程中,需要将风电功率不确定性因素考虑进来。因此,利用机会约束规划方法来建立包含风电场在内的日前经济调度随机优化模型,并选用改进的引力搜索算法来求解所建立的模型。最后,选用具体算例,结合IEEE-30节点网络调试系统进行模型仿真,并利用仿真结果探讨了负荷水平对系统所容纳的风电容量的影响、风电功率预测偏差对预调度惩罚成本的影响、风电预测功率下不同成本系数对日前市场的影响。研究结果表明,利用改进引力搜索算法进行优化仿真的收敛速度比粒子群算法和遗传算法的收敛速度快,可靠性高。研究所得成果为日前市场经济下风电功率短期预测研究提供参考。
Considering the random fluctuation characteristics of wind power itself, the uncertainty factors of wind power are necessary to be taken into account in the process of making pre-day market economic pre-scheduling. Therefore, a pre-day economic scheduling stochastic optimization model consisting of wind farm is established herein with the chance constrained programming method, and then the improved gravitational search algorithm is selected to solve the established model. Finally, the model simulation is carried out with specific calculation case in combination with the IEEE-30 node network debugging system, while the influence from the load level on the system containing wind power capacity, the influence from the predictive deviation of wind power on the penalty cost of pre-scheduling and the influences from different cost coefficients under predicted wind powerare discussed with the simulated result. The study result shows that the rate of convergence optimally simulated by the improved gravitational search algorithm is faster than those from the particle swarm optimization and genetic algorithm with high reliability. The study result provides reference for the study on the short-term prediction of wind power under pre-day market economy.
Considering the random fluctuation characteristics of wind power itself, the uncertainty factors of wind power are necessary to be taken into account in the process of making pre-day market economic pre-scheduling. Therefore, a pre-day economic scheduling stochastic optimization model consisting of wind farm is established herein with the chance constrained programming method, and then the improved gravitational search algorithm is selected to solve the established model. Finally, the model simulation is carried out with specific calculation case in combination with the IEEE-30 node network debugging system, while the influence from the load level on the system containing wind power capacity, the influence from the predictive deviation of wind power on the penalty cost of pre-scheduling and the influences from different cost coefficients under predicted wind powerare discussed with the simulated result. The study result shows that the rate of convergence optimally simulated by the improved gravitational search algorithm is faster than those from the particle swarm optimization and genetic algorithm with high reliability. The study result provides reference for the study on the short-term prediction of wind power under pre-day market economy.
引文
[1] 谭忠富,鞠立伟.中国风电发展综述:历史、现状、趋势及政策[J].华北电力大学学报(社会科学版),2013(2):1- 7.
[2] 王玫,赵晓丽.中国风电发展经济政策回顾与分析[J].中国能源,2011,33(10):10- 14.
[3] 黄德琥,陈继军,张岚.大规模风电并网对电力系统的影响[J].广东电力,2010,23(7):27- 30.
[4] 刘瑞叶,黄磊.基于动态神经网络的风电场输出功率预测[J].电力系统自动化,2012,36(11):19- 22.
[5] 董根源,温步瀛,江岳文.考虑发电机报价曲线的发电计划偏差的β-PSO优化方法研究[J].电力自动化设备,2006,26(7):18- 21.
[6] 张伯明,吴文传,郑太一,等.消纳大规模风电的多时间尺度协调的有功调度系统设计[J].电力系统自动化,2011,35(1):1- 6.
[7] 吴栋梁,王扬,郭创新,等.电力市场环境下考虑风电预测误差的经济调度模型[J].电力系统自动化,2012,36(6):23- 28.
[8] HETZER J, YU D C, BHATTARAI K. An economic dispatch model incorporating wind power[J].IEEE Transactions on Energy Conversion,2008,23(2):603- 611.
[9] 袁铁江,晁勤,吐尔逊.伊不拉音,等.面向电力市场的含风电电力系统的环境经济调度优化[J].电网技术,2009,33(20):131- 135.
[10] 江岳文,陈冲,温步瀛.基于随机模拟粒子群算法的含风电场电力系统经济调度[J].电工电能新技术,2007,26(3):37- 41.
[11] 江岳文,陈冲,温步瀛.含风电场的电力系统机组组合问题随机模拟粒子群算法[J].电工技术学报,2009,24(6):129- 137.
[12] 黄永皓,尚金成,康重庆,等.电力日前交易市场的运作机制及模型[J].电力系统自动化,2003,27(3):23- 27.
[13] 周玮, 孙辉, 顾宏,等. 计及风险备用约束的含风电场电力系统动态经济调度[J]. 中国电机工程学报, 2012, 32(1):47- 55.
[14] RASHEDI E,NEZAMABADI-POUR H,SAR YYAZDI S.GSA: a gravitational search algorithm[J].Information Sciences, 2009,179(13): 2232- 2248.
[15] 王颖帆, 余代海, 江岳文. 利用贝叶斯学习法修正报价的发电权交易研究[J]. 广东电力, 2016(1):36- 40.
[2] 王玫,赵晓丽.中国风电发展经济政策回顾与分析[J].中国能源,2011,33(10):10- 14.
[3] 黄德琥,陈继军,张岚.大规模风电并网对电力系统的影响[J].广东电力,2010,23(7):27- 30.
[4] 刘瑞叶,黄磊.基于动态神经网络的风电场输出功率预测[J].电力系统自动化,2012,36(11):19- 22.
[5] 董根源,温步瀛,江岳文.考虑发电机报价曲线的发电计划偏差的β-PSO优化方法研究[J].电力自动化设备,2006,26(7):18- 21.
[6] 张伯明,吴文传,郑太一,等.消纳大规模风电的多时间尺度协调的有功调度系统设计[J].电力系统自动化,2011,35(1):1- 6.
[7] 吴栋梁,王扬,郭创新,等.电力市场环境下考虑风电预测误差的经济调度模型[J].电力系统自动化,2012,36(6):23- 28.
[8] HETZER J, YU D C, BHATTARAI K. An economic dispatch model incorporating wind power[J].IEEE Transactions on Energy Conversion,2008,23(2):603- 611.
[9] 袁铁江,晁勤,吐尔逊.伊不拉音,等.面向电力市场的含风电电力系统的环境经济调度优化[J].电网技术,2009,33(20):131- 135.
[10] 江岳文,陈冲,温步瀛.基于随机模拟粒子群算法的含风电场电力系统经济调度[J].电工电能新技术,2007,26(3):37- 41.
[11] 江岳文,陈冲,温步瀛.含风电场的电力系统机组组合问题随机模拟粒子群算法[J].电工技术学报,2009,24(6):129- 137.
[12] 黄永皓,尚金成,康重庆,等.电力日前交易市场的运作机制及模型[J].电力系统自动化,2003,27(3):23- 27.
[13] 周玮, 孙辉, 顾宏,等. 计及风险备用约束的含风电场电力系统动态经济调度[J]. 中国电机工程学报, 2012, 32(1):47- 55.
[14] RASHEDI E,NEZAMABADI-POUR H,SAR YYAZDI S.GSA: a gravitational search algorithm[J].Information Sciences, 2009,179(13): 2232- 2248.
[15] 王颖帆, 余代海, 江岳文. 利用贝叶斯学习法修正报价的发电权交易研究[J]. 广东电力, 2016(1):36- 40.