为了更加准确有效地对极地光伏发电功率做出预测,提出一种基于GA-GNNM的极地光伏发电功率预测方法。首先对采集到的气候因素数据以及光伏发电数据中缺失、异常部分进行清洗归一化处理;通过最大相关最小冗余算法（MRMR）选择最佳的气候特征组合,构建多维气候特征数据集;并将其输入到K均值聚类算法中完成不同季节天气类型聚类划分,然后引入相对距离量化寒暖季不同天气类型下与预测日相似度高的发电功率;构建灰色神经融合模型（GNNM）,将灰色模型（GM）的微分方程映射入神经网络模型,并采用遗传优化算法（GA）对模型参数进行优化以避免局部最优,提高极地光伏预测算法的精度。最后,以南极恩克斯堡岛气候及光伏发电数据为例进行验证,算例分析的结果为极地第5个新站的建立奠定理论基础。
Abstract
Photovoltaic power forecasting is one of the basics for optimal configuration and coordinated operation of the microgrid. However, due to the extreme conditions of the two seasons in the polar region： polar day and night, and more stormy snow, occasional snow fog weather etc., the polar photovoltaic power is greatly affected, with instability, uncertainty and non-linear characteristics. In order to forecasting the power of polar photovoltaic more accurately and effectively, a polar photovoltaic power forecasting method based on GA-GNNM is proposed. First, cleaning and normalizing abnormal and missing data, useing the maximum correlation minimum redundancy algorithm to select the best feature set, the K-means clustering algorithm is used to cluster the weather types of different seasons, and the relative distance is used to quantify the power with high similarity to the predicted day under different weather types. Construct a gray neural hybrid model （GNNM）, map the differential equations of the gray model into the neural network model, and use genetic optimization algorithm （GA） to optimize the model parameters to avoid local optimization and improve the accuracy of the polar energy prediction algorithm. Finally, the climate and photovoltaic power data of Enksburg Island in Antarctica are used as examples for verification. The results of the example analysis lay a theoretical foundation for the establishment of the fifth new station in the polar region.
关键词
光伏发电 /
K均值算法 /
特征选择 /
预测分析 /
灰色神经融合模型
Key words
photovoltaic power generation /
K-means algorithms /
feature selection /
predictive analysis /
grey neural network modal
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 曹俊波, 周任军, 邓学华, 等. 考虑优化ARIMA模型差分次数的风功率预测[J]. 电力系统及其自动化学报, 2019, 31(1): 105-111.
CAO J B, ZHOU R J, DENG X H, et al. Wind power forecast considering differential times of optimal ARIMA model[J]. Proceedings of the CSU-EPSA, 2019, 31(1): 105-111.
[2] 丁明, 张超, 王勃, 等. 基于功率波动过程的风电功率短期预测及误差修正[J]. 电力系统自动化, 2019, 43(2): 2-9.
DING M, ZHANG C, WANG B, et al. Short-term forecasting and error correction of wind power based on power fluctuation process[J]. Automation of electric power systems, 2019, 43(2): 2-9.
[3] 于浩, 秦文萍, 魏斌, 等. 考虑预测误差的交直流混合微电网经济调度策略[J]. 电网技术, 2019, 43(11): 3987-3996.
YU H, QIN W P, WEI B, et al. Economic dispatch Strategy of AC-DC hybrid microgrid considering forecast errors[J]. Power system technology, 2019, 43(11): 3987-3996.
[4] 单英浩, 付青, 耿炫.基于改进BP-SVM-ELM与粒子化SOM-LSF的微电网光伏发电组合预测方法[J]. 中国电机工程学报, 2016, 36(12): 3334-3342.
SHAN Y H, FU Q, GENG X.Combined forecasting of photovoltaic power generation in microgrid based on the improved BP-SVM-ELM and SOM-LSF with particlization[J]. Proceedings of the CSEE, 2016, 36(12): 3334-3342.
[5] UTPAL K D, KOK S T, MEHDI S, et al. Forecasting of photovoltaic power generation and model optimization:A review[J]. Photovoltaic and sustainable energy reviews, 2018, 81(1): 912-928.
[6] 王小杨, 罗多, 孙韵琳, 等. 基于ABC-SVM和PSO-RF的光伏微电网日发电功率组合预测方法研究[J]. 太阳能学报, 2020, 41(3): 177-183.
WANG X Y, LUO D, SUN Y L, et al. Research on daily combined power forecasting method of photovoltaic microgrid based on ABC-SVM and PSO-RF[J]. Acta energiae solaris sinica, 2020, 41(3): 177-183.
[7] 谢从珍, 王江储, 谢心昊, 等. 基于细粒度特征的BOA-GBDT光伏出力预测[J]. 电网技术, 2020, 44(2): 689-696.
XIE C Z, WANG J C, XIE X H, et al. BOA-GBDT photovoltaic output prediction based on fine-grained features[J]. Power system technology, 2020, 44(2): 689-696.
[8] 张倩, 马愿, 李国丽, 等. 频域分解和深度学习算法在短期负荷及光伏功率预测中的应用[J]. 中国电机工程学报, 2019, 39(8): 2221-2230, 5.
ZHANG Q, MA Y, LI G L, et al. Application of frequency domain decomposition and deep learning algorithm in short-term load and photovoltaic power prediction[J]. Proceedings of the CSEE, 2019, 39(8): 2221-2230,5.
[9] 赵康宁, 蒲天骄, 王新迎, 等. 基于改进贝叶斯神经网络的光伏出力概率预测[J]. 电网技术, 2019, 43(12): 4377-4386.
ZHAO K N, PU T J, WANG X Y, et al. Probability prediction of photovoltaic output based on improved bayesian neural network[J]. Power system technology, 2019, 43(12): 4377-4386.
[10] 王育飞, 付玉超, 薛花.计及太阳辐射和混沌特征提取的光伏发电功率DMCS-WNN预测法[J]. 中国电机工程学报, 2019, 39(S1): 63-71.
WANG Y F, FU Y C, XUE H.DMCS-WNN prediction method of photovoltaic power generation considering solar radiation and chaos feature extraction[J]. Proceedings of the CSEE, 2019, 39(S1): 63-71.
[11] 王守相, 张娜.基于灰色神经网络组合模型的光伏短期出力预测[J]. 电力系统自动化, 2012, 36(19): 37-41.
WANG S X, ZHANG N.Prediction of PV short-term output based on grey neural network combination model[J]. Power system automation, 2012, 36(19): 37-41.
[12] 宋人杰, 刘福盛, 马冬梅.基于相似日和WNN的光伏发电功率超短期预测模型[J]. 电测与仪表, 2017, 54(7): 75-80.
SONG R J, LIU F S, MA D M.Ultra-short-term prediction model of photovoltaic power generation based on similar days and WNN[J]. Electrical measurement and instrumentation, 2017, 54(7): 75-80.
[13] 于秋玲, 许长清, 李珊.基于模糊聚类和支持向量机的短期光伏功率预测[J]. 电力系统及其自动化学报, 2016, 28(12): 115-118, 129.
YU Q L, XU C Q, LI S.Short-term photovoltaic power prediction based on fuzzy clustering and support vector machine[J]. Journal of power system and automation, 2016, 28(12): 115-118,129.
[14] 王新普, 周想凌, 邢杰.一种基于改进灰色BP神经网络组合的光伏出力预测方法[J]. 电力系统保护与控制, 2016, 44(18): 81-87.
WANG X P, ZHOU X L, XING J.A photovoltaic output prediction method based on improved gray BP neural network combination[J]. Power system protection and control, 2016, 44(18): 81-87.
[15] 章勇高, 王妍, 孙佳, 等. 基于灰色神经网络优化组合的风力发电量预测研究[J]. 电测与仪表, 2014, 51(22): 30-34.
ZHANG Y G, WANG Y, SUN J, et al. Research on wind power generation forecast based on optimized combination of gray neural network[J]. Electric measurement & instrumentation, 2014, 51(22): 30-34.
基金
上海市“科技创新行动计划”（19DZ1207604）; 国家自然科学基金（51977127）
PDF(1825 KB)
