一种计及空间相关性的光伏电站历史出力数据的修正方法

doi:10.6040/j.issn.1672-3961.0.2020.431

摘要/Abstract

摘要： 针对光伏系统渗透率增高对电力系统稳定运行带来的严峻挑战,考虑到光伏功率预测技术精度高度依赖于数据精度的问题,提出一种基于人工神经网络的光伏电站历史出力数据修正方法。利用人工神经网络在建立复杂非线性映射关系的优越性,引入皮尔逊相关系数对数据进行降维处理,选择与目标光伏电站出力相关性高的电站作为基准光伏电站,并结合光伏出力的空间相关性特征与基准光伏电站的出力数据对目标光伏电站失准及缺失数据进行修正,以解决由人为因素或数据采集系统老旧带来的光伏数据失准问题,并通过山东省聊城市的光伏历史出力数据对所提方法进行分析验证。

关键词: 光伏发电, 数据降维, 功率预测, 人工神经网络, 数据修正

Abstract: The increase of photovoltaic(PV)system penetration rate brought great challenges to the stable operation of power system. Considering that the accuracy of photovoltaic power prediction was highly dependent on data accuracy, this paper proposed a correction method for historical output data of photovoltaic power plant by taking the advantages of strong ability of artificial neural network in mapping complex nonlinear relations. Person correlation coefficient was employed to select reference PV plants for dimensionality reduction. The inaccurate and missing data of photovoltaic power station could be identified and corrected by taking the spatial correlation characteristics of PV output into consideration based on the output power of reference PV plants, which could solve the problem of PV data inaccuracy caused by human factors or data acquisition system aging. The proposed method was analyzed and verified by the historical output data of Liaocheng City in Shandong Prvoince.

Key words: photovoltaic power, dimensionality reduction, power forecasting, artificial neural network, data correction

中图分类号:

TM615

尹晓敏,孟祥剑,侯昆明,陈亚潇,高峰. 一种计及空间相关性的光伏电站历史出力数据的修正方法[J]. 山东大学学报 (工学版), 2021, 51(4): 118-123.

YIN Xiaomin, MENG Xiangjian, HOU Kunming, CHEN Yaxiao, GAO Feng. Correction method for historical output data of photovoltaic power plant considering spatial correlation based on artificial neural network[J]. Journal of Shandong University(Engineering Science), 2021, 51(4): 118-123.

参考文献

[1] 周海,李登宣,尹万思,等.极限学习机的光伏发电短期预测校正方法[J].电网与清洁能源,2020,36(6):64-69. ZHOU Hai, LI Dengxuan, YIN Wansi, et al. Short-term forecasting correction method of photovoltaic power based on extreme learning machine[J]. Power System and Clean Energy, 2020, 36(6):64-69.
[2] SHENG Hanmin, XIAO Jian, CHENG Yuhua, et al. Short-term solar power forecasting based on weighted gaussian process regression[J]. IEEE Transactions on Industrial Electronics, 2018, 65(1):300-308.
[3] 汪海瑛,白晓民.并网光伏的短期运行备用评估[J].电力系统自动化,2013,37(5):55-60. WANG Haiying, BAI Xiaomin. Short-term operating reserve assessment for grid-connected photovoltaic system[J]. Automation of Electric Power Systems, 2013, 37(5):55-60.
[4] 龚莺飞,鲁宗相,乔颖,等.光伏功率预测技术[J].电力系统自动化,2016,40(4):140-151. GONG Yingfei, LU Zongxiang, QIAO Ying, et al. An overview of photovoltaic energy system output forecasting technology[J]. Automation of Electric Power Systems, 2016, 40(4):140-151.
[5] 时珉,周海,韩雨彤,等. 一种考虑季节特性的光伏电站多模型功率预测方法[J].电网与清洁能源, 2019, 35(7):75-82. SHI Min, ZHOU Hai, HAN Yutong, et al. A multi-model power forecasting approach of photovoltaic plant based on seasonal characteristics[J]. Power System and Clean Energy, 2019, 35(7):75-82.
[6] HUANG Chioujye, KUO Pinghuan. Multiple-input deep convolutional neural network model for short-term photovoltaic power forecasting[J]. IEEE Access, 2019, 7: 74822-74834.
[7] 李正明,高赵亮,梁彩霞. 基于FCM和CG-DBN的光伏功率短期预测[J]. 现代电力, 2019, 36(5): 62-67. LI Zhengming, GAO Zhaoliang, LIANG Caixia. Short-term prediction of photovoltaic power based on combination of FCM and CG-DBN[J]. Modern Electric Power, 2019, 36(5): 62-67.
[8] 张华彬,杨明玉.基于最小二乘支持向量机的光伏出力超短期预测[J]. 现代电力, 2015, 32(1): 70-75. ZHANG Huabin, YANG Mingyu. Ultra short-term forecasting for photovoltaic power putput based on least square support vector machine[J]. Modern Electric Power, 2015, 32(1): 70-75.
[9] 栗然,李广敏.基于支持向量机回归的光伏发电出力预测 [J].中国电力,2008,41(2):74-78. LI Ran, LI Guangmin. Photovoltaic power generation output forecasting based on support vector machine regression technique[J]. Electric Power, 2008, 41(2):74-78.
[10] 李光明,刘祖明,何京鸿,等. 基于多元线性回归模型的并网光伏发电系统发电量预测研究[J]. 现代电力, 2011, 28(2): 43-48. LI Guangming, LIU Zuming, HE Jinghong, et al. Study on the generator forecasting of grid-connected PV power system based on multivariate linear regression model[J]. Modern Electric Power, 2011, 28(2): 43-48.
[11] KUSHWAHA Vishal, PINDORIYA Naran M. Very short-term solar PV generation forecast using SARIMA model: a case study[C] //2017 7th International Conference on Power Systems(ICPS).Pune, India: IEEE, 2017.
[12] SMARAGDIS Paris, RAJ Bhiksha, SHASHANKA Madhusudana. Missing data imputation for spectral audio signals[C] //IEEE International Workshop on Machine Learning for Signal Processing.Grenoble, France: IEEE, 2009.
[13] DEMPSTER A, LAIRD N, RUBIN D. Maximum likelihood form incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society Series B(Methodological), 1977, 39(1): 1-38.
[14] TANAKA Takafumi, KAWAKAMI Wataru, KUWABARA Seiki, et al. A. Intelligent monitoring of optical fiber bend using artificial neural networks trained with constellation data[J]. IEEE Networking Lett, 2019, 1(2):60-62.
[15] CHATTERJEE P, KARAN B, SINHA P. Multi-layer feed-forward modular network for induction motor[C] //Proceedings of the 2002 International Joint Conference on Neural Networks. Onolulu, USA: IEEE, 2002.
[16] 喻成林,黄强,李钢. 观测站缺失数据修复的神经网络模型研究[J]. 矿山测量,2014, 1(1):92-95. YU Chenglin, HUANG Qiang, LI Gang. Research on neural network model of missing data repair[J]. Mine Survey, 2014, 1(1):92-95.
[17] 张一平,宋红,罗敏,等.人工神经网络算法在光伏发电短期功率预测中的应用[J].太阳能,2019(4):47-53. ZHANG Yiping, SONG Hong, LUO Min, et al. Application of artificial neural network algorithm in short-term power prediction of PV power generation[J]. Solar Energy, 2019(4):47-53.
[18] 杨茂,周宜. 计及风电场状态的风电功率超短期预测[J]. 中国电机工程学报, 2019, 39(5):1259-1267. YANG Mao, ZHOU Yi. Ultra-short-term prediction of wind power cconsidering wind farm status[J]. Proceeding of the CESS, 2019, 39(5):1259-1267.
[19] 卜凡鹏,陈俊艺,张琪祁,等.一种基于双层迭代聚类分析的负荷模式可控精细化识别方法[J].电网技术,2018,42(3):903-913. BU Fanpeng, CHEN Junyi, ZHANG Qiqi, et al. A control refined recognition method of electrical load pattern based on bilayer iterative clustering analysis[J]. Power System Technology, 2018, 42(3):903-913.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed