漂浮式光伏网格对海上天气突变的感知方法

doi:10.16183/j.cnki.jsjtu.2021.526

摘要/Abstract

摘要：

漂浮式光伏在海洋上的应用目前主要受制于海底电缆与特质浮筒的成本问题,该能量如果被海上牧场等场景的无人管理系统就地消纳则表现出高度的适用匹配性.这种场景下由漂浮式光伏形成的网格系统可以解决海上牧场等对突变天气的预警需求.由于光伏出力模型对天气随机变化的强跟随性, 基于大面积光伏的时空相关性分析,通过硬件、距离、时延、天气等因素建立相似电站融合估计关系.基于长短期记忆网络算法对相似电站时序进行跟踪,所得超短期预测值可以估计目标相似电站的状态.用某市城区规模的数据验证了该思路的可行性,表明该框架可以弥补传统研究的不足.

关键词: 漂浮式光伏, 海洋天气, 相似电站, 长短期记忆网络

Abstract:

Currently, the application of floating photovoltaics in the ocean is mainly restricted by the cost of submarine cables and special buoys. It will show a high degree of applicability if the energy is consumed by the unmanned management systems on ocean farms and in other scenarios. The grid system formed by the floating photovoltaics can satisfy the early warning requirements of the sudden weather changes on ocean farms. Due to the strong follow-up of the photovoltaic output model to random weather changes, based on the spatial-temporal correlation analysis of large-area photovoltaics, hardware, distance, time delay, and weather, a similar power station fusion estimation relationship is established. Based on the long short-term memory (LSTM) algorithm, the ultra-short-term prediction value of the time sequence tracking of similar power stations can be used to estimate the early warning of the status of target similar power stations. The city-scale data was used to verify the feasibility of the proposed idea, which shows that the framework can complement traditional research deficiencies.

Key words: floating photovoltaic, marine weather, similar power station, long short-term memory (LSTM)

中图分类号:

TM615

姜淏予, 王沛伦, 葛泉波, 徐今强, 罗朋, 姚刚. 漂浮式光伏网格对海上天气突变的感知方法[J]. 上海交通大学学报, 2022, 56(12): 1584-1597.

JIANG Haoyu, WANG Peilun, GE Quanbo, XU Jinqiang, LUO Peng, YAO Gang. A Sensing Method Based of Floating Photovoltaic Grids to Sudden Changes in Marine Weather[J]. Journal of Shanghai Jiao Tong University, 2022, 56(12): 1584-1597.

图/表 21

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

表1

表2

图15

图16

图17

图18

表3

参考文献 24

[1]	HALLEGRAEFF G M. Ocean climate change, phytoplankton community responses, and harmful algal blooms: A formidable predictive challenge[J]. Journal of Phycology, 2010, 46(2): 220-235. doi: 10.1111/j.1529-8817.2010.00815.x URL
[2]	黄强, 郭怿, 江建华, 等. “双碳”目标下中国清洁电力发展路径[J]. 上海交通大学学报, 2021, 55(12): 1499-1509.
	HUANG Qiang, GUO Yi, JIANG Jianhua, et al. Development pathway of China’s clean electricity under carbon peaking and carbon neutrality goals[J]. Journal of Shanghai Jiao Tong University, 2021, 55(12): 1499-1509.
[3]	SPENCER R S, MACKNICK J, AZNAR A, et al. Floating photovoltaic systems: Assessing the technical potential of photovoltaic systems on man-made water bodies in the Continental United States[J]. Environmental Science & Technology, 2019, 53(3): 1680-1689. doi: 10.1021/acs.est.8b04735 URL
[4]	王浤宇, 王佩明, 李艳红, 等. 水上漂浮式光伏发电系统[J]. 华电技术, 2017, 39(3): 74-76.
	WANG Hongyu, WANG Peiming, LI Yanhong, et al. Floating photovoltaic power generation system on water[J]. Huadian Technology, 2017, 39(3): 74-76.
[5]	孙祖峰, 陈佩杭. 漂浮式光伏应用及技术难点简析[J]. 科技创新与应用, 2016 (12): 37-38.
	SUN Zufeng, CHEN Peihang. Brief analysis of floating photovoltaic application and technical difficulties[J]. Science and Technology Innovation and Application, 2016 (12): 37-38.
[6]	TANG R, LIN Q, ZHOU J, et al. Suppression strategy of short-term and long-term environmental disturbances for maritime photovoltaic system[J]. Applied Energy, 2020, 259: 114183. doi: 10.1016/j.apenergy.2019.114183 URL
[7]	ZHOU Y, ZHOU N, GONG L, et al. Prediction of photovoltaic power output based on similar day analysis, genetic algorithm and extreme learning machine[J]. Energy, 2020, 204: 117894. doi: 10.1016/j.energy.2020.117894 URL
[8]	祝暄懿, 姚李孝. 基于相似日和小波神经网络的光伏短期功率预测[J]. 电网与清洁能源, 2019, 35(3): 75-78.
	ZHU Xuanyi, YAO Lixiao. Solar power plant short-term power forecast based on similar days and WNN[J]. Advances of Power System & Hydroelectric Engineering, 2019, 35(3): 75-78.
[9]	焦田利, 章坚民, 李熊, 等. 基于空间相关性的大规模分布式用户光伏空间分群方法[J]. 电力系统自动化, 2019, 43(21): 97-105.
	JIAO Tianli, ZHANG Jianmin, LI Xiong, et al. Spatial clustering method for large-scale distributed user photovoltaics based on spatial correlation[J]. Automation of Electric Power Systems, 2019, 43(21): 97-105.
[10]	LI Y, HAN D, YAN Z. Long-term system load forecasting based on data-driven linear clustering method[J]. Journal of Modern Power Systems and Clean Energy, 2018, 6(2), 306-316. doi: 10.1007/s40565-017-0288-x URL
[11]	STANISLAWSKI B, MARGAIRAZ F, CAL R B, et al. Potential of module arrangements to enhance convective cooling in solar photovoltaic arrays[J]. Renewable Energy, 2020, 157: 851-858. doi: 10.1016/j.renene.2020.04.049 URL
[12]	郭珂, 袁丹夫, 温浚铎, 等. 新型光伏阵列等效串联电阻的计算模型[J]. 中国测试, 2019, 45(12): 126-131.
	GUO Ke, YUAN Danfu, WEN Junduo, et al. Calculation model of equivalent series resistance of new photovoltaic array[J]. China Measurement and Testing Technology, 2019, 45(12): 126-131.
[13]	何啸, 李国富, 葛霞, 等. 漂浮式光伏发电装置在海浪影响下的光照性能研究[J]. 工程设计学报, 2014, 21(6): 545-549.
	HE Xiao, LI Guofu, GE Xia, et al. Analysis of irradiation on floating solar panels affected by ocean waves[J]. Journal of Engineering Design, 2014, 21(6): 545-549. doi: 10.1080/09544820802353297 URL
[14]	张商州, 楚冰清, 袁训锋, 等. 光伏阵列模型分析及最大功率点跟踪研究[J]. 自动化与仪器仪表, 2019(10): 114-116.
	ZHANG Shangzhou, CHU Bingqing, YUAN Xunfeng, et al. Photovoltaic array model analysis and maximum power point tracking study[J]. Automation & Instrumentation, 2019(10): 114-116.
[15]	FRNDA J, DURICA M, NEDOMA J, et al. A weather forecast model accuracy analysis and ECMWF enhancement proposal by neural network[J]. Sensors (Basel, Switzerland), 2019, 19(23): 31771275.
[16]	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.
[17]	ZHU X. Prediction of short-term PV output power based on PCA-stacking under different weather conditions[J]. IOP Conference Series: Materials Science and Engineering, 2020, 853(1): 012023. doi: 10.1088/1757-899X/853/1/012023 URL
[18]	ZHOU R, ZHANG X. Application and practice of high-precision solar resource monitoring technology in photovoltaic power plant area[J]. Journal of Power and Energy Engineering, 2019, 7(11): 22-30.
[19]	郭辉, 杨国清, 姚李孝, 等. 基于综合相似日和功率相关性的光伏电站预测功率修正[J]. 电网与清洁能源, 2018, 34(9): 52-58.
	GUO Hui, YANG Guoqing, YAO Lixiao, et al. Correction of predictive power of pv plants based on integrated similar days and power correlations[J]. Advances of Power System and Hydroelectric Engineering, 2018, 34(9): 52-58.
[20]	荆博, 谭伦农, 钱政, 等. 光伏发电短期预测研究进展综述[J]. 电测与仪表, 2017, 54(12): 1-6.
	JING Bo, TAN Lunong, QIAN Zheng, et al. An overview of research progress of short-term photovoltaic forecasts[J]. Electrical Measurement & Instrumentation, 2017, 54(12): 1-6.
[21]	彭旭. 基于时序生产模拟的新能源消纳评估方法研究[D]. 西安: 西安理工大学, 2019.
	PENG Xu. Research on evaluation method of new energy absorption based on time series production simulation[D]. Xi’an: Xi’an University of Technology, 2019.
[22]	CHAUHAN A, TYAGI V V, ANAND S. Minimum entropy generation and its validation against Hottel Whillier model for PV/T and FPC collectors[J]. Solar Energy, 2019, 188: 143-157. doi: 10.1016/j.solener.2019.05.063 URL
[23]	DONYA A, MOHAMMADD A, MAJID B. A robust method for early diagnosis of autism spectrum disorder from EEG signals based on feature selection and DBSCAN method[J]. Biocybernetics and Biomedical Engineering, 2020, 40(1): 482-493. doi: 10.1016/j.bbe.2020.01.008 URL
[24]	JUHYOK U, LU P Y, KIM C S, et al. A new LSTM based reversal point prediction method using upward/downward reversal point feature sets[J]. Chaos, Solitons & Fractals, 2020, 132: 109559. doi: 10.1016/j.chaos.2019.109559 URL

聚类段	欧氏距离	电站标号
聚类段1	0.7680	13
聚类段2	0.8523	8
聚类段3	0.8739	26
聚类段4	0.9262	14

聚类段	改进距离	欧氏距离	电站标号
聚类段1	0.0299	0.0332	16
聚类段2	0.0428	0.0475	14
聚类段3	0.0582	0.0647	4

时间	RMSE		误差/%
时间	优化前	优化后	优化前	优化后
第5天	0.8491	0.7694	6.54	5.93
第13天	1.0242	0.7821	7.63	6.00
第16天	3.1840	2.4114	23.14	14.35
第28天	1.3802	1.2580	8.72	8.38