[1]张 洁,郝倩男.基于烟花算法优化 BP 神经网络的光伏功率预测[J].计算机技术与发展,2021,31(10):146-153.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 025]
　ZHANG Jie,HAO Qian-nan.Forecast of Photovoltaic Power Generation Based on Firework Algorithm Optimized BP Neural Network[J].,2021,31(10):146-153.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 025]
点击复制

基于烟花算法优化 BP 神经网络的光伏功率预测()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年10期
页码:
146-153
栏目:
应用前沿与综合
出版日期:
2021-10-10

文章信息/Info

Title:
Forecast of Photovoltaic Power Generation Based on Firework Algorithm Optimized BP Neural Network
文章编号:
1673-629X(2021)10-0146-08
作者:
张 洁; 郝倩男
南京邮电大学 计算机学院,江苏 南京 210023
Author(s):
ZHANG Jie; HAO Qian-nan
School of Computer Science and Technology,Nanjing University of Posts and Telecommunications,Nanjing 210023,China
关键词:
光伏发电; 功率预测; 烟花算法; 遗传算法; BP 神经网络
Keywords:
photovoltaic power generation; power prediction; firework algorithm; genetic algorithm; BP neural network
分类号:
TP301. 6
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 10. 025
摘要:
光伏发电功率受到气象状况及环境因素的影响,具有很强的随机性与间歇性,给电网的安全运行带来了一系列的问题。 为了准确预测光伏发电功率,降低光伏并网的不利影响,实现电网系统的稳定运行,将烟花算法( FWA) 引入到神经网络模型中,利用烟花算法局部搜索能力和全局搜索能力自调节的机制优化神经网络中权重和阈值的寻优过程,提出了一种基于烟花算法改进 BP 神经网络( FWA-BP)的光伏发电功率预测模型。 通过某光伏电站实测数据进行仿真实验,结果表明,与传统 BP 神经网络和遗传算法优化 BP 神经网络相比,FWA-BP 神经网络模型的预测结果更接近实际值,均方根误差、平均绝对百分比等误差指标更低,模型更稳定,从而说明该方法可以快速准确地实现对光伏发电功率的预测。
Abstract:
Photovoltaic power generation is affected by meteorological conditions and environmental factors with strong randomness and intermittentness, which brings a series of problems to the safe operation of the power grid. In order to accurately predict photovoltaic power generation,reduce the adverse effects of photovoltaic grid-connected and achieve stable operation of the grid system, the firework algorithm ( FWA) is introduced into? the neural network model. The optimization process of weights and thresholds in neural network is optimized by the self-adjusting mechanism of local search ability and global search ability of fireworks algorithm. An improved BP neural network ( FWA-BP) based on the firework algorithm is proposed to predict the photovoltaic power generation power. Simulation experiments are carried out through the measured data of a photovoltaic power station,and it is showed that compared with traditional BP neural network and genetic algorithm optimized BP neural network,the prediction result of FWA-BP neural network model is closer to the actual value,the error indicators such as root mean square error and average absolute percentage are lower,and the model is more stable. The proposed method can quickly and accurately realize the prediction of photovoltaic power generation.

相似文献/References:

[1]彭 飞,张健男,张晓华,等.基于初始化非负矩阵分解的光伏发电预测[J].计算机技术与发展,2021,31(02):185.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 034]
　PENG Fei,ZHANG Jian-nan,ZHANG Xiao-hua,et al.Prediction of Photovoltaic Power Generation Based on Initialized Non-negative Matrix Factorization[J].,2021,31(10):185.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 034]
[2]张 洁,罗俊杰.基于 TS-FOA 算法的光伏组件参数辨识[J].计算机技术与发展,2022,32(06):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 003]
　ZHANG Jie,LUO Jun-jie.Parameter Identification of Photovoltaic Modules Based on TS-FOA[J].,2022,32(10):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 003]

更新日期/Last Update: 2021-10-10