基于数据增强和优化DHKELM的短期光伏功率预测

郭利进; 马粽阳; 胡晓岩

doi:10.19912/j.0254-0096.tynxb.2024-0693

PDF(7175 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (8) : 463-471. DOI: 10.19912/j.0254-0096.tynxb.2024-0693

基于数据增强和优化DHKELM的短期光伏功率预测

郭利进^1,2, 马粽阳^1,2, 胡晓岩^1,2

作者信息 +

SHORT-TERM PHOTOVOLTAIC POWER PREDICTION BASED ON DATA AUGMENTATION AND OPTIMIZATION OF DHKELM

Guo Lijin^1,2, Ma Zongyang^1,2, Hu Xiaoyan^1,2

Author information +

文章历史 +

摘要

针对不同气象条件数据质量差异较大且光伏功率呈高波动性难以预测等问题,提出添加随机噪声的数据增强方法（DA）和改进的神经网络组合模型。首先利用谱聚类算法将光伏数据按不同气象条件进行分类,随后通过添加与输入同形状的随机噪声方法提升数据集的规模与质量。针对深度混合核极限学习机（DHKELM）超参数多等问题,提出融合佳点集初始化、黄金正弦更新策略、非线性扰动和最优个体自适应扰动的改进鹈鹕优化算法（IPOA）对其超参数寻优。最后以青海共和县光伏园内某电站数据为例,结果表明基于数据增强的改进鹈鹕算法优化深度混合核极限学习机（DA-IPOA-DHKELM）模型在不同天气、季节条件下预测误差最小,拟合度均能达到90%以上,改进模型预测精度高、算法适用性强。

Abstract

Addressing the issues of significant differences in data quality under different meteorological conditions and the high volatility in photovoltaic power which make it difficult to predict, a combined model is proposed that incorporates data augmentation （DA） and optimized deep hybrid kernel extreme learning machine （DHKELM）. At the outset, the spectral clustering algorithm is applied to categorize photovoltaic data based on varying meteorological conditions. Subsequently, the data set is expanded and its quality is enhanced by adding random noise of the same shape as the input data. Considering the numerous hyperparameters of DHKELM, a multi-strategy improved IPOA is proposed that integrates good point set initialization, golden sine update strategy, nonlinear perturbation, and adaptive perturbation of the optimal individual for hyperparameter optimization. Employing data from a photovoltaic station in Gonghe county photovoltaic park, Qinghai as a case study, the results demonstrate that the DA-IPOA-DHKELM model minimizes prediction errors under different weather and seasonal conditions and achieves fitting accuracy exceeding 90%, significantly enhancing the precision of photovoltaic power predictions and the applicability of the algorithm.

导出引用

郭利进, 马粽阳, 胡晓岩. 基于数据增强和优化DHKELM的短期光伏功率预测[J]. 太阳能学报. 2025, 46(8): 463-471 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0693

Guo Lijin, Ma Zongyang, Hu Xiaoyan. SHORT-TERM PHOTOVOLTAIC POWER PREDICTION BASED ON DATA AUGMENTATION AND OPTIMIZATION OF DHKELM[J]. Acta Energiae Solaris Sinica. 2025, 46(8): 463-471 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0693

中图分类号： TM615

参考文献

[1] 叶林, 路朋, 赵永宁, 等. 含风电电力系统有功功率模型预测控制方法综述[J]. 中国电机工程学报, 2021, 41(18): 6181-6198.
YE L, LU P, ZHAO Y N, et al.Review of model predictive control for power system with large-scale wind power grid-connected[J]. Proceedings of the CSEE, 2021, 41(18): 6181-6198.
[2] LI Q, ZHANG X Y, MA T J, et al.A multi-step ahead photovoltaic power prediction model based on similar day, enhanced colliding bodies optimization, variational mode decomposition, and deep extreme learning machine[J]. Energy, 2021, 224: 120094.
[3] MAYER M J, YANG D Z.Pairing ensemble numerical weather prediction with ensemble physical model chain for probabilistic photovoltaic power forecasting[J]. Renewable and sustainable energy reviews, 2023, 175: 113171.
[4] LI Y T, HE Y, SU Y, et al.Forecasting the daily power output of a grid-connected photovoltaic system based on multivariate adaptive regression splines[J]. Applied energy, 2016, 180: 392-401.
[5] GHIMIRE S, DEO R C, RAJ N, et al.Deep solar radiation forecasting with convolutional neural network and long short-term memory network algorithms[J]. Applied energy, 2019, 253: 113541.
[6] 薛阳, 燕宇铖, 贾巍, 等. 基于改进灰狼算法优化长短期记忆网络的光伏功率预测[J]. 太阳能学报, 2023, 44(7): 207-213.
XUE Y, YAN Y C, JIA W, et al.Photovoltaic power prediction model based on IGWO-LSTM[J]. Acta energiae solaris sinica, 2023, 44(7): 207-213.
[7] 焦丕华, 蔡旭, 王乐乐, 等. 考虑数据分解和进化捕食策略的BiLSTM短期光伏发电功率预测[J]. 太阳能学报, 2024, 45(2): 435-442.
JIAO P H, CAI X, WANG L L, et al.BiLSTM short-term photovoltaic power prediction considering data decomposition and evolutionary predation strategies[J]. Acta energiae solaris sinica, 2024, 45(2): 435-442.
[8] 王超, 蔺红, 庞晓虹. 基于HPO-VMD和MISMA-DHKELM的短期光伏功率组合预测[J]. 太阳能学报, 2023, 44(12): 65-73.
WANG C, LIN H, PANG X H.Short-term photovoltaic power combination prediction based on HPO-VMD and MISMA-DHKELM[J]. Acta energiae solaris sinica, 2023, 44(12): 65-73.
[9] MORENO-BAREA F J, JEREZ J M, FRANCO L. Improving classification accuracy using data augmentation on small data sets[J]. Expert systems with applications, 2020, 161: 113696.
[10] CHU Y F, FENG C Y, GUO C L, et al.Network embedding based on deep extreme learning machine[J]. International journal of machine learning and cybernetics, 2019, 10(10): 2709-2724.
[11] NIE Y H, ZAMZAM A S, BRANDT A.Resampling and data augmentation for short-term PV output prediction based on an imbalanced sky images dataset using convolutional neural networks[J]. Solar energy, 2021, 224: 341-354.
[12] TROJOVSKÝ P, DEHGHANI M.Pelican optimization algorithm: a novel nature-inspired algorithm for engineering applications[J]. Sensors, 2022, 22(3): 855.
[13] 冯增喜, 李嘉乐, 葛珣, 等. 融合多策略改进鲸鱼优化算法及其应用[J]. 计算机集成制造系统, 2025, 31(2): 590-603.
FENG Z X, LI J L, GE X, et al.Integrating multi-strategy improved whale optimization algorithm and its application[J]. Computer integrated manufacturing systems, 2025, 31(2): 590-603.
[14] 张帅, 王俊杰, 李爱莲, 等. 集成正态云和动态扰动的哈里斯鹰优化算法[J]. 小型微型计算机系统, 2023, 44(6): 1153-1161.
ZHANG S, WANG J J, LI A L, et al.Harris hawk optimization algorithm integrating normal clouds and dynamic perturbations[J]. Journal of Chinese computer systems, 2023, 44(6): 1153-1161.
[15] TANYILDIZI, E, DEMIR G.Golden sine algorithm: a novel math-inspired algorithm[J]. Advances in electrical and computer engineering, 17(2): 71-78.
[16] 刘成汉, 何庆. 融合多策略的黄金正弦黑猩猩优化算法[J]. 自动化学报, 2023, 49(11): 2360-2373.
LIU C H, HE Q.Golden sine chimp optimization algorithm integrating multiple strategies[J]. Acta automatica sinica, 2023, 49(11): 2360-2373.
[17] 朱学敏, 刘升, 朱学林, 等. 多策略混合改进的海洋捕食者算法及其工程应用[J]. 国外电子测量技术, 2023, 42(5): 125-134.
ZHU X M, LIU S, ZHU X L, et al.Enhancing marine predator algorithm with hybrid multi-strategy and its application[J].Foreign electronic measurement technology, 2023, 42(5): 125-134.
[18] MARINI F, WALCZAK B.Particle swarm optimization (PSO). a tutorial[J]. Chemometrics and intelligent laboratory systems, 2015, 149: 153-165.
[19] CHOPRA N, MOHSIN ANSARI M.Golden jackal optimization: a novel nature-inspired optimizer for engineering applications[J]. Expert systems with applications, 2022, 198: 116924.
[20] RAO R V.Jaya: a simple and new optimization algorithm or solving constrained and unconstrained optimization problems[J]. International journal of industrial engineering computations, 2016, 7(1): 19-34.
[21] ZHAO S J, ZHANG T R, MA S L, et al.Sea-horse optimizer: a novel nature-inspired meta-heuristic for global optimization problems[J]. Applied intelligence, 2023, 53(10): 11833-11860.
[22] 商立群, 黄辰浩, 侯亚东, 等. 采用特征优选和优化深层核极限学习机的短期风电功率预测[J]. 西安交通大学学报, 2023, 57(1): 66-77.
SHANG L Q, HUANG C H, HOU Y D, et al.Short-term wind power prediction by using the deep kernel extreme learning machine with well-selected and optimized features[J]. Journal of Xi’an Jiaotong University, 2023, 57(1): 66-77.
[23] 奉国和. SVM分类核函数及参数选择比较[J]. 计算机工程与应用, 2011, 47(3): 123-124, 128.
FENG G H.Parameter optimizing for support vector machines classification[J]. Computer engineering and applications, 2011, 47(3): 123-124, 128.