针对现有的非均匀风电场发电功率预测方法准确度不高、可解释性差等问题,提出一种改进的物理引导的图神经网络（IPGNN）模型来预测非均匀风电场的发电功率输出。首先,构建一个基于三维高斯尾流模型的物理信息基函数,可更准确地反映非均匀风电场中风力机之间尾流相互作用的关系;其次,设计一组基于消息传递框架的图神经网络更新策略,该策略在图神经网络边缘更新中结合注意力机制将物理信息基函数作为权重更新函数,可增强模型的可解释性。在不同数量风力机下的非均匀风电场仿真实验表明,相比于基于一维尾流模型的PGNN模型和典型的数据驱动模型,IPGNN模型均可获得较好的预测效果,其中对具有20台风力机的非均匀风电场,发电功率预测平均绝对误差为3.92%,可认为是一种有效的非均匀风电场发电功率预测方法。
Abstract
Aiming at the problems of low accuracy and poor interpretability of the existing power prediction methods of non-uniform wind farm layout, we propose an improved physics-induced graph neural network （IPGNN） model to predict the power output of non-uniform wind farms. Firstly, IPGNN constructs a physical information basis function based on three-dimensional Gaussian wake model, which can more accurately reflect the wake interaction between wind turbines in non-uniform wind farms; Secondly, IPGNN designs a set of graph neural network update strategies based on the message passing framework. In the graph neural network edge update strategy, combined with the attention mechanism, the physical information basis function is used as the weight update function, which enhances the interpretability of the model. The simulation experiments of non-uniform wind farms with different numbers of wind turbines show that IPGNN model has a good prediction effect compared with PGNN model based on one-dimensional wake model and typical data-driven model. For non-uniform wind farms with 20 wind turbines, the MAPE of power generation prediction is 3.92%. It is an effective method for power generation prediction of non-uniform wind farm layout.
关键词
深度学习 /
图神经网络 /
风电功率 /
风电场 /
数值模拟
Key words
deep learning /
graph neural network /
wind power /
wind farm layout /
numerical simulation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] WISER R, LANTZ E, MAI T, et al.Wind vision: a new era for wind power in the United States[J]. The electricity journal, 2015, 28(9): 120-132.
[2] Global Wind Energy Council. GWEC global wind report[R]. Brussels, Belgium: Global Wind Energy Council, 2022. 1-148.
[3] BARTHELMIE R J, HANSEN K, FRANDSEN S T, et al.Modelling and measuring flow and wind turbine wakes in large wind farms offshore[J]. Wind energy, 2009, 12(5): 431-444.
[4] CHEN Y, LI H, JIN K, et al.Wind farm layout optimization using genetic algorithm with different hub height wind turbines[J]. Energy conversion and management, 2013, 70: 56-65.
[5] MOSETTI G, POLONI C, DIVIACCO B.Optimization of wind turbine positioning in large windfarms by means of a genetic algorithm[J]. Journal of wind engineering and industrial aerodynamics, 1994, 51(1): 105-116.
[6] HOU P, HU W H, SOLTANI M, et al.Optimized placement of wind turbines in large-scale offshore wind farm using particle swarm optimization algorithm[J]. IEEE transactions on sustainable energy, 2015, 6(4): 1272-1282.
[7] WANG Y, LIU H, LONG H, et al.Differential evolution with a new encoding mechanism for optimizing wind farm layout[J]. IEEE transactions on industrial informatics, 2017, 14(3): 1040-1054.
[8] 袁桂丽, 吴振民, 刘骅骐, 等. 基于深度置信网络的短期风电功率预测[J]. 太阳能学报, 2022, 43(2): 451-457.
YUAN G L, WU Z M, LIU H Q, et al.Short-term wind power prediction based on deep belief network[J]. Acta energiae solaris sinica, 2022, 43(2): 451-457.
[9] 苏向敬, 周汶鑫, 李超杰, 等. 基于双重注意力LSTM神经网络的可解释海上风电出力预测[J]. 电力系统自动化, 2022, 46(7): 141-151.
SU X J, ZHOU W X, LI C J, et al.Interpretable offshore wind power output forecasting based on long short-term memory neural network with dual-stage attention[J]. Automation of electric power systems, 2022, 46(7): 141-151.
[10] PARK J Y, PARK J K.Physics-induced graph neural network: an application to wind-farm power estimation[J]. Energy, 2019, 187(1): 115883-115897.
[11] JENSEN N O.A note on wind generator interaction[M]. Roskilde, Denmark: Riso National Laboratory, 1983.
[12] TAO S Y, XU Q S, FEIJÓO A, et al. Bi-hierarchy optimization of a wind farm considering environmental impact[J]. IEEE transactions on sustainable energy, 2020, 11(4): 2515-2524.
[13] 宋翌蕾, 田琳琳, 赵宁. 风力机三维尾流模型的提出与校核[J]. 太阳能学报, 2021, 42(2): 129-135.
SONG Y L, TIAN L L, ZHAO N.Proposal and validation of a new 3D wake model for wind turbine[J]. Acta energiae solaris sinica, 2021, 42(2): 129-135.
[14] SUN H Y, YANG H X.Numerical investigation of the average wind speed of a single wind turbine and development of a novel three-dimensional multiple wind turbine wake model[J]. Renewable energy, 2020, 147: 192-203.
[15] FLEMING P, KING J, BAY C J, et al.Overview of FLORIS updates[J]. Journal of physics: conference series, 2020, 1618(2): 022028.
[16] KING J, FLEMING P, KING R, et al.Control-oriented model for secondary effects of wake steering[J]. Wind energy science, 2021, 6(3): 701-714.
[17] FARRELL A, KING J, DRAXL C, et al.Design and analysis of a wake model for spatially heterogeneous flow[J]. Wind energy science, 2021, 6(3): 737-758.
[18] GUO X J, ZHANG Y, YAN J T, et al.Integrated dynamics response analysis for IEA 10-MW spar floating offshore wind turbine[J]. Journal of marine science and engineering, 2022, 10(4): 542-559.
[19] STANLEY A P J, NING A. Coupled wind turbine design and layout optimization with nonhomogeneous wind turbines[J]. Wind energy science, 2019, 4(1): 99-114.
基金
国家自然科学基金（62366012; 61563012）; 广西自然科学基金（2021GXNSFAA220074）; 广西嵌入式技术与智能系统重点实验室基金 （2020-1-3）
PDF(2276 KB)
