基于人工蜂群算法改进BP神经网络的光伏发电量预测
|
摘要
新能源产业发展对电网电能消纳提出新的要求和挑战,准确的发电量预测对电力调度部门合理转供、优化分配负荷起着重要作用.光伏发电量受到天气、云层遮挡、温湿度的影响较大,发电量变化快、数据量大,针对传统BP神经网络负荷预测方法的不足,提出了基于人工蜂群改进的ABC-BP神经网络算法,融合两种算法在全局迭代、局部搜索能力以及泛化能力方面的优势,整合训练样本天气、温度等影响因素,对光伏出力进行预测.仿真结果表明,改进型的人工蜂群BP神经网络较传统BP神经网络算法具有准确性高、均方误差小的优点,同时收敛速度加快,具有更好的稳定性和鲁棒性,可以有效预测发电量的变化,为电力调度提供更为准确的参考,有较强的实践应用价值.
The development of new energy industry puts forward new requirements and challenges for the power consumption of power grid,and accurate power generation prediction plays an important role in the rational transfer of power supply and optimal load distribution by power dispatching departments.Photovoltaic power generation is greatly affected by weather,cloud cover,temperature and humidity,and the power generation changes rapidly and the amount of data is large.In view of the deficiency of traditional BP neural network load forecasting methods,an improved ABC-BP neural network algorithm based on artificial bee colony is proposed in this paper,which integrates the advantages of the two algorithms in global iteration,local search ability and generalization ability,integrating the influence factors such as weather and temperature of training samples,and forecasting photovoltaic power.The simulation results show that the improved artificial bee colony BP neural network has the advantages of high accuracy and small mean square error compared with the traditional BP neural network algorithm.At the same time,it has faster convergence speed,better stability and robustness,which can effectively predict the change of power generation and provide a more accurate reference for power dispatching.
The development of new energy industry puts forward new requirements and challenges for the power consumption of power grid,and accurate power generation prediction plays an important role in the rational transfer of power supply and optimal load distribution by power dispatching departments.Photovoltaic power generation is greatly affected by weather,cloud cover,temperature and humidity,and the power generation changes rapidly and the amount of data is large.In view of the deficiency of traditional BP neural network load forecasting methods,an improved ABC-BP neural network algorithm based on artificial bee colony is proposed in this paper,which integrates the advantages of the two algorithms in global iteration,local search ability and generalization ability,integrating the influence factors such as weather and temperature of training samples,and forecasting photovoltaic power.The simulation results show that the improved artificial bee colony BP neural network has the advantages of high accuracy and small mean square error compared with the traditional BP neural network algorithm.At the same time,it has faster convergence speed,better stability and robustness,which can effectively predict the change of power generation and provide a more accurate reference for power dispatching.
引文
[1]丁明,王磊,毕锐.基于改进BP神经网络的光伏发电系统输出功率短期预测模型[J].电力系统保护与控制,2012,40(11):93-99.
[2]谭子兵,黄秀超,钟建伟.基于BP人工神经网络的电力变压器故障诊断研究[J].湖北民族学院学报(自然科学版),2018,36(1):89-92.
[3]王莹,程薇薇.基于神经网络算法预测水解酶的亚类[J].赤峰学院学报(自然科学版),2018,34(4).45-46.
[4]张鹏华.基于神经网络的分布式光伏电站短期功率预测[D].淮南:安徽理工大学,2017.
[5]杨柳宁,孙以泽,孟婥,等.基于改进的BP神经网络的光伏组件发电量预测模型设计[J].水电能源科学,2013,31(9):243-246.
[6]苏彩红,向娜,陈广义.基于人工蜂群算法与BP神经网络的水质评价模型[J].环境工程学报,2012,6(2):699-704.
[7]杨咪,徐盼盼,钱会,等.基于人工蜂群算法的BP双隐含层神经网络水质模型[J].环境监测管理与技术,2018,30(1):21-26.
[8]KARABOGA D,OZTURK C.Neural networks training by artificial bee colony algorithm on pattern classification[J].Neural Network World,2009,19(3):279-292.
[9]KARABOGA D,BASTURK B.A powerful and efficient algorithm for numerical function optimization:artificial bee colony(ABC)algorithm[J].Journal of Global Optimization,2007,39(3):459-471.
[10]FRISCH K V.The dance language and orientation of bees[M].Cambridge,Massachusetts:Belknap Press of Harvard University Press,1967.
[11]TERESHKO V,LOENGAROV A.Collective decision-making in honey bee foraging dynamics[J].Computing and Information Systems Journal,2005,9(3):1-7.
[12]秦全德,程适,李丽,等.人工蜂群算法研究综述[J].智能系统报,2014,9(2):127-135.
[13]郭庆春,郝源,李雪,等.BP神经网络在长江水质COD预测中的应用[J].计算机技术与发展,2014,24(4):235-238.
[2]谭子兵,黄秀超,钟建伟.基于BP人工神经网络的电力变压器故障诊断研究[J].湖北民族学院学报(自然科学版),2018,36(1):89-92.
[3]王莹,程薇薇.基于神经网络算法预测水解酶的亚类[J].赤峰学院学报(自然科学版),2018,34(4).45-46.
[4]张鹏华.基于神经网络的分布式光伏电站短期功率预测[D].淮南:安徽理工大学,2017.
[5]杨柳宁,孙以泽,孟婥,等.基于改进的BP神经网络的光伏组件发电量预测模型设计[J].水电能源科学,2013,31(9):243-246.
[6]苏彩红,向娜,陈广义.基于人工蜂群算法与BP神经网络的水质评价模型[J].环境工程学报,2012,6(2):699-704.
[7]杨咪,徐盼盼,钱会,等.基于人工蜂群算法的BP双隐含层神经网络水质模型[J].环境监测管理与技术,2018,30(1):21-26.
[8]KARABOGA D,OZTURK C.Neural networks training by artificial bee colony algorithm on pattern classification[J].Neural Network World,2009,19(3):279-292.
[9]KARABOGA D,BASTURK B.A powerful and efficient algorithm for numerical function optimization:artificial bee colony(ABC)algorithm[J].Journal of Global Optimization,2007,39(3):459-471.
[10]FRISCH K V.The dance language and orientation of bees[M].Cambridge,Massachusetts:Belknap Press of Harvard University Press,1967.
[11]TERESHKO V,LOENGAROV A.Collective decision-making in honey bee foraging dynamics[J].Computing and Information Systems Journal,2005,9(3):1-7.
[12]秦全德,程适,李丽,等.人工蜂群算法研究综述[J].智能系统报,2014,9(2):127-135.
[13]郭庆春,郝源,李雪,等.BP神经网络在长江水质COD预测中的应用[J].计算机技术与发展,2014,24(4):235-238.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |