基于特征选择和XGBoost算法考虑极端天文、气象事件影响的光伏性能预测方法

王瑶; 吴云来; 俞铁铭; 胡华友; 李明光

doi:10.19912/j.0254-0096.tynxb.2023-0014

PDF(2094 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (5) : 547-555. DOI: 10.19912/j.0254-0096.tynxb.2023-0014

基于特征选择和XGBoost算法考虑极端天文、气象事件影响的光伏性能预测方法

王瑶¹, 吴云来², 俞铁铭², 胡华友¹, 李明光¹

作者信息 +

FORECASTING METHOD OF PHOTOVOLTAIC POWER GENERATION BASED ON FEATURE SELECTION AND XGBOOST ALGORITHM CONSIDERING INFLUENCE OF EXTREME ASTRONOMICAL AND METEOROLOGICAL EVENTS

Wang Yao¹, Wu Yunlai², Yu Tieming², Hu Huayou¹, Li Mingguang¹

Author information +

文章历史 +

摘要

以Pearson’s r特征选择方法进行参数相关性判断后,构建分析决策树指数模型,提高水平面总辐射的预测精度,通过气象参数的主成分提取,实现训练集降维。采用XGBoost算法构建预测模型,加入正则项控制模型的复杂度,降低过拟合率,提高模型对未知数据的适应能力。通过泰勒展开将损失函数的选取和算法优化过程去耦合,实现极端天文、气象条件下光伏电站性能的预测和模型评估。预测结果与实测值对比表明,所提预测法能自动学习缺失值的处理策略,支持多种类型的基分类器,有广泛的优化空间。在针对光伏功率P_w、系统效率PR、产能利用率CF的预测平均绝对百分比误差在15%以内,显示出良好的预测准确度和稳定性。

Abstract

After the parameter correlation was judged by Pearson's r feature selection method, the analysis decision tree index model was built to improve the prediction accuracy of the total horizontal radiation intensity, and the dimensionality reduction of the training set was realized through the principal component extraction of meteorological parameters. XGBoost algorithm is used to construct the prediction model, adding regular terms to control the complexity of the model, reducing the overfitting rate and improving the adaptability of the model to unknown data. By means of Taylor expansion, the selection of loss function and algorithm optimization process are decoupted, and the performance prediction and model evaluation of photovoltaic power stations under extreme astronomical and meteorological conditions are realized. The comparison between the predicted results and the measured values shows that the proposed method can automatically learn the missing value processing strategy, support various types of base classifiers, and have a wide range of optimization space. This model has good prediction accuracy and stability, as the average absolute percentage error in predicting photovoltaic power P_w, Performance ratio（PR）, and capacity utilization factor（CF）is within 15%.

导出引用

王瑶, 吴云来, 俞铁铭, 胡华友, 李明光. 基于特征选择和XGBoost算法考虑极端天文、气象事件影响的光伏性能预测方法[J]. 太阳能学报. 2024, 45(5): 547-555 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0014

Wang Yao, Wu Yunlai, Yu Tieming, Hu Huayou, Li Mingguang. FORECASTING METHOD OF PHOTOVOLTAIC POWER GENERATION BASED ON FEATURE SELECTION AND XGBOOST ALGORITHM CONSIDERING INFLUENCE OF EXTREME ASTRONOMICAL AND METEOROLOGICAL EVENTS[J]. Acta Energiae Solaris Sinica. 2024, 45(5): 547-555 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0014

中图分类号： TM615

参考文献

[1] 何必涛. 气象预报及预警技术在新能源发电领域中的研究及运用[J]. 工程技术, 2021(7): 164-165.
HE B T.Research and application of meteorological forecasting and early warning technology in the field of new energy power generation[J]. Engineering technology, 2021(7): 164-165
[2] 王倩, 诺桑, 措加旺姆, 等. 西藏日食太阳辐射观测研究[J]. 光谱学与光谱分析, 2021, 41(12): 3892-3900.
WANG Q, NORSANG G, TSOJA W, et al.Solar radiation observation during a solar eclipse in Tibet[J]. Spectroscopy and spectral analysis, 2021, 41(12): 3892-3900.
[3] 梁志峰, 秦放, 崔方. “6·21”日环食对光伏发电及电网运行影响分析[J]. 电力系统自动化, 2021, 45(7): 1-7.
LIANG Z F, QIN F, CUI F.Impact analysis of annular solar eclipse on June 21, 2020 in China on photovoltaic power generation and power grid operation[J]. Automation of electric power systems, 2021, 45(7): 1-7.
[4] 鞠冠章, 王靖然, 崔琛, 等. 极端天气事件对新能源发电和电网运行影响研究[J]. 智慧电力, 2022, 50(11): 77-83.
JU G Z, WANG J R, CUI C, et al.Impact of extreme weather events on new energy power generation and power grid operation[J]. Smart power, 2022, 50(11): 77-83.
[5] 靳双龙, 迟永宁, 王勃, 等. 气象传感监测技术在电网中的应用研究[J]. 电力信息与通信技术, 2020, 18(4): 84-90.
JIN S L, CHI Y N, WANG B, et al.Application research of meteorological sensing and monitoring technology in power grid[J]. Electric power information and communication technology, 2020, 18(4): 84-90.
[6] 刘纯, 马烁, 董存, 等. 欧洲3·20日食对含大规模光伏发电的电网运行影响及启示[J]. 电网技术, 2015, 39(7): 1765-1772.
LIU C, MA S, DONG C, et al.Review and experiences of the European solar eclipse’s impact on power grid operation with significant PV generation[J]. Power system technology, 2015, 39(7): 1765-1772.
[7] 周滢垭, 廖宇. 2015年3月20日日食对德国电网的影响[J]. 南方电网技术, 2015, 9(5): 15-18.
ZHOU Y Y, LIAO Y.Influence of solar eclipse on March 20^th, 2015 on German power grid[J]. Southern power system technology, 2015, 9(5): 15-18.
[8] LIBRA M, BERÁNEK V, SEDLÁČEK J, et al. Roof photovoltaic power plant operation during the solar eclipse[J]. Solar energy, 2016, 140: 109-112.
[9] MADHAVAN B L, VENKAT RATNAM M.Impact of a solar eclipse on surface radiation and photovoltaic energy[J]. Solar energy, 2021, 223: 351-366.
[10] 刘昶. 考虑极端天气因素的并网型光伏出力研究[D]. 北京: 华北电力大学, 2018.
LIU C.Rerseach on grid-connected photovoltaic output considering extreme weather factors[D]. Beijing: North China Electric Power University, 2018.
[11] 宋智林. 光伏发电系统的短期发电功率预测研究[D]. 天津: 天津大学, 2016.
SONG Z L.Study on short-term power forecasting of PV power system[D]. Tianjin: Tianjin University, 2016.
[12] 秦东东. 光伏功率预测及其在微电网中的应用[D]. 杭州: 浙江大学, 2016.
QIN D D.Photovoltaic power forecasting and its application in micro grid[D]. Hangzhou: Zhejiang University, 2016.
[13] LITJENS G B M A, WORRELL E, VAN SARK W G J H M. Assessment of forecasting methods on performance of photovoltaic-battery systems[J]. Applied energy, 2018, 221: 358-373.
[14] 崔杨, 陈正洪, 孙朋杰. 弃光限电条件下不同纬度地区短期光伏发电功率预测对比分析[J]. 太阳能学报, 2018, 39(6): 1610-1618.
CUI Y, CHEN Z H, SUN P J.Comparison and analysis of short-term PV power prediction at different latitude under condition of discarding PV power[J]. Acta energiae solaris sinica, 2018, 39(6): 1610-1618.
[15] ACHLEITNER S, KAMTHE A, LIU T, et al.SIPS: solar irradiance prediction system[C]//IPSN-14 Proceedings of the 13th International Symposium on Information Processing in Sensor Networks. Berlin, Germany, 2014: 225-236.
[16] 朱红路, 李旭, 姚建曦, 等. 基于小波分析与神经网络的光伏电站功率预测方法[J]. 太阳能学报, 2015, 36(11): 2725-2730.
ZHU H L, LI X, YAO J X, et al.The power prediction method for photovoltaic power station based on wavelet analysis and neural networks[J]. Acta energiae solaris sinica, 2015, 36(11): 2725-2730.
[17] LI Y T, SU Y, SHU L J.An ARMAX model for forecasting the power output of a grid connected photovoltaic system[J]. Renewable energy, 2014, 66: 78-89.
[18] SHI H, YUAN X T, ZHANG J H, et al.Adaptive photovoltaic power computer prediction technology using deep learning and LSSVM[J]. Journal of physics: conference series, 2021, 2033(1): 012072.
[19] 高正. 基于改进神经网络的光伏发电功率短期预测研究[D]. 长春: 长春工业大学, 2019.
GAO Z.Short term prediction of photovoltaic power generation based on improved neural network[D]. Changchun: Changchun University of Technology, 2019.
[20] 蒋晋文, 刘伟光. XGBoost算法在制造业质量预测中的应用[J]. 智能计算机与应用, 2017, 7(6): 58-60.
JIANG J W, LIU W G.Application of XGBoost algorithm in manufacturing quality prediction[J]. Intelligent computer and applications, 2017, 7(6): 58-60.
[21] 赵舫, 盛青, 王新刚, 等. 基于主成分分析合作博弈的电网负荷智能预测方法研究[J]. 电网与清洁能源, 2022, 38(2): 48-52, 60.
ZHAO F, SHENG Q, WANG X G, et al.A study on the intelligent load forecasting method of power grids based on principal component analysis and cooperative game[J]. Power system and clean energy, 2022, 38(2): 48-52, 60.
[22] 张峰, 宋晓娜, 薛惠锋, 等. 水资源消耗预测的异常值检测及缺失数据填补方法[J]. 统计与决策, 2018, 34(16): 13-17.
ZHANG F, SONG X N, XUE H F, et al.Method for abnormal data detection and missing data filling in water resources consumption forecasting[J]. Statistics & decision, 2018, 34(16): 13-17.
[23] WANG Y P, ZHAO J J, YANG C N, et al.Remaining useful life prediction of rolling bearings based on Pearson correlation-KPCA multi-feature fusion[J]. Measurement, 2022, 201: 111572.
[24] 张博文, 闫龙, 陆凌辉, 等. 光伏电站汇集系统单极接地故障的皮尔逊相关系数识别方法[J]. 电力系统及其自动化学报, 2022, 34(2): 116-121.
ZHANG B W, YAN L, LU L H, et al.Single-pole grounding fault identification method for photovoltaic power station collection system based on Pearson correlation coefficient[J]. Proceedings of the CSU-EPSA, 2022, 34(2): 116-121.
[25] 张棪, 曹健. 面向大数据分析的决策树算法[J]. 计算机科学, 2016, 43(S1): 374-379, 383.
ZHANG Y, CAO J.Decision tree algorithms for big data analysis[J]. Computer science, 2016, 43(S1): 374-379, 383.
[26] 季一木, 张永潘, 郎贤波, 等. 面向流数据的决策树分类算法并行化[J]. 计算机研究与发展, 2017, 54(9): 1945-1957.
JI Y M, ZHANG Y P, LANG X B, et al.Parallel of decision tree classification algorithm for stream data[J]. Journal of computer research and development, 2017, 54(9): 1945-1957.
[27] 丁坤, 刘振飞, 高列, 等. 基于主成分分析和马氏距离的光伏系统健康状态研究[J]. 可再生能源, 2017, 35(1): 1-7.
DING K, LIU Z F, GAO L, et al.Research on photovoltaic system health state based on PCA-MD method[J]. Renewable energy resources, 2017, 35(1): 1-7.
[28] 卓助航, 杨俊杰, 胡丰晔, 等. 基于XGBoost短期预测控制的微电网多时间尺度调度[J]. 计算机应用与软件, 2021, 38(5): 54-60.
ZHUO Z H, YANG J J, HU F Y, et al.Multi-time scale scheduling of microgrid based on XGBoost short-term predictive control[J]. Computer applications and software, 2021, 38(5): 54-60.
[29] 罗预欣, 张兵, 薛运强. 基于变量分析和粒子群优化加权随机森林的交通事件检测方法[J]. 科学技术与工程, 2021, 21(14): 6044-6049.
LUO Y X, ZHANG B, XUE Y Q.Weighted random forest based on variable analysis and particle swarm optimization traffic incident detection method[J]. Science technology and engineering, 2021, 21(14): 6044-6049.
[30] 王灵矫, 夏伟, 谭貌, 等. 供电系统电力短期负荷预测优化仿真研究[J]. 计算机仿真, 2018, 35(4): 53-57.
WANG L J, XIA W, TAN M, et al.Power system short-term load forecasting optimization simulation research[J]. Computer simulation, 2018, 35(4): 53-57.
[31] 牟宏, 王春义, 顾洁, 等. 基于支持向量-分位数回归的短期负荷预测及风险分析[J]. 电力科学与技术学报, 2016, 31(3): 51-56.
MOU H, WANG C Y, GU J, et al.Short-term load forecasting and risk analysis based on support vector-quantile regression[J]. Journal of electric power science and technology, 2016, 31(3): 51-56.