基于KPCA和HC的IPSO-LSTM光伏出力预测模型研究

徐昌; 许野; 王晓晖; 孟亦康; 秦宇; 李薇

doi:10.19912/j.0254-0096.tynxb.2024-0159

PDF(3869 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (5) : 362-374. DOI: 10.19912/j.0254-0096.tynxb.2024-0159

基于KPCA和HC的IPSO-LSTM光伏出力预测模型研究

徐昌¹, 许野¹, 王晓晖², 孟亦康¹, 秦宇¹, 李薇¹

作者信息 +

RESEARCH ON COMBINED PHOTOVOLTAIC OUTPUT PREDICTION MODEL BASED ON KPCA, HC AND IPSO-LSTM

Xu Chang¹, Xu Ye¹, Wang Xiaohui², Meng Yikang¹, Qin Yu¹, Li Wei¹

Author information +

文章历史 +

摘要

为提升光伏发电功率的预测精度,构建一套融合核主成分分析法（KPCA）、层次聚类（HC）算法、改进粒子群算法（IPSO）和长短期记忆神经网络（LSTM）的光伏出力组合预测模型。在运用KPCA方法对影响光伏出力的气象因素进行降维和生成主成分因子的基础上,联合使用HC算法和综合相似距离法挑选出与待预测日气象要素匹配度较高且内部耦合性强的历史日样本集,并运用IPSO优化生成LSTM神经网络的最优超参数组合,最终实现云南某光伏电站发电量的精准预测。对比其他模型,所提组合预测方法在不同天气类型下均能实现较好的预测效果,具有广阔的应用前景。

Abstract

To improve the prediction accuracy of photovoltaic （PV） power generation, a combined PV output prediction model is established by incorporating kernel principal component analysis （KPCA）, hierarchical clustering （HC）, improved particle swarm optimization （IPSO） and long short-term memory （LSTM） into a general framework. The KPCA method is firstly used to reduce the dimension of meteorological factors affecting photovoltaic output and generate their principal component factors. Then, the HC algorithm and comprehensive similarity distance method are employed jointly to select historical similar days with the high internal correlation and similar meteorological characteristics to the predicted day. Next, the hyperparameter combination of LSTM neural network is optimized by the IPSO algorithm for establishing high-precision prediction model. Finally, the power generation of a photovoltaic plant in Yunnan province is accurately estimated. Compared with other prediction models, the combined forecasting method proposed in this study achieves better forecasting results under various weather types and has the broad application prospects.

导出引用

徐昌, 许野, 王晓晖, 孟亦康, 秦宇, 李薇. 基于KPCA和HC的IPSO-LSTM光伏出力预测模型研究[J]. 太阳能学报. 2025, 46(5): 362-374 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0159

Xu Chang, Xu Ye, Wang Xiaohui, Meng Yikang, Qin Yu, Li Wei. RESEARCH ON COMBINED PHOTOVOLTAIC OUTPUT PREDICTION MODEL BASED ON KPCA, HC AND IPSO-LSTM[J]. Acta Energiae Solaris Sinica. 2025, 46(5): 362-374 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0159

中图分类号： TM615

参考文献

[1] 叶林, 裴铭, 路朋, 等. 基于天气分型的短期光伏功率组合预测方法[J]. 电力系统自动化, 2021, 45(1): 44-54.
YE L, PEI M, LU P, et al.Combination forecasting method of short-term photovoltaic power based on weather classification[J]. Automation of electric power systems, 2021, 45(1): 44-54.
[2] LIMOUNI T, YAAGOUBI R, BOUZIANE K, et al.Accurate one step and multistep forecasting of very short-term PV power using LSTM-TCN model[J]. Renewable energy, 2023, 205: 1010-1024.
[3] ZHU J B, LI M R, LUO L, et al.Short-term PV power forecast methodology based on multi-scale fluctuation characteristics extraction[J]. Renewable energy, 2023, 208: 141-151.
[4] 张雲钦, 程起泽, 蒋文杰, 等. 基于EMD-PCA-LSTM的光伏功率预测模型[J]. 太阳能学报, 2021, 42(9): 62-69.
ZHANG Y Q, CHENG Q Z, JIANG W J, et al.Photovoltaic power prediction model based on EMD-PCA-LSTM[J]. Acta energiae solaris sinica, 2021, 42(9): 62-69.
[5] GUPTA P, SINGH R.Combining a deep learning model with multivariate empirical mode decomposition for hourly global horizontal irradiance forecasting[J]. Renewable energy, 2023, 206: 908-927.
[6] LI X H, JIA R C, ZHANG R R, et al.A KPCA-BRANN based data-driven approach to model corrosion degradation of subsea oil pipelines[J]. Reliability engineering & system safety, 2022, 219: 108231.
[7] 王利福, 魏玉琪, 刘屹江泽. 基于KPCA与IPFA-KELM的变压器故障识别方法[J]. 控制工程, 2023, 30(7): 1180-1189.
WANG L F, WEI Y Q, LIU Y.Transformer fault identification method based on KPCA and IPFA-KELM[J]. Control engineering of China, 2023, 30(7): 1180-1189.
[8] ZHI Y, SUN T, YANG X D.A physical model with meteorological forecasting for hourly rooftop photovoltaic power prediction[J]. Journal of building engineering, 2023, 75: 106997.
[9] 常青松, 杨昭, 杨熠辉, 等. 基于相似日聚类的超短期光伏功率组合预测模型[J]. 热力发电, 2023, 52(11): 123-131.
CHANG Q S, YANG Z, YANG Y H.Ultrashort term photovoltaic power combinatorial forecasting model based on similar day clustering[J]. Thermal power generation, 2023, 52(11): 123-131.
[10] 王开艳, 杜浩东, 贾嵘, 等. 基于相似日聚类和QR-CNN-BiLSTM模型的光伏功率短期区间概率预测[J]. 高电压技术, 2022, 48(11): 4372-4388.
WANG K Y, DU H D, JIA R, et al.Short-term interval probability prediction of photovoltaic power based on similar daily clustering and QR-CNN-BiLSTM model[J]. High voltage engineering, 2022, 48(11): 4372-4388.
[11] 董雪, 赵宏伟, 赵生校, 等. 基于SOM聚类和二次分解的BiGRU超短期光伏功率预测[J]. 太阳能学报, 2022, 43(11): 85-93.
DONG X, ZHAO H W, ZHAO S X, et al.Ultra-short-term forecasting method of photovoltaic power based on SOM clustering, secondary decomposition and BiGRU[J]. Acta energiae solaris sinica, 2022, 43(11): 85-93.
[12] CHEN W H, CHENG L S, CHANG Z P, et al.Interval prediction of photovoltaic power using improved NARX network and density peak clustering based on kernel mahalanobis distance[J]. Complexity, 2022, 2022(1): 8169510.
[13] ZHANG Y Y, XUE P, ZHAO Y F, et al.Spectra measurement and clustering analysis of global horizontal irradiance for solar energy application[J]. Renewable energy, 2024, 222: 119813.
[14] 李超然, 潘鹏程, 杨伟荣等. 基于改进相似日优化HBA-BiLSTM-KELM的光伏发电功率预测[J]. 太阳能学报, 2024, 45(5): 508-516.
LI C R, PAN P C, YANG W R, et al.Research on PV system power prediction based on improved similar day and HBA-BiLSTM-KELM neural netword[J]. Acta energiae solaris sinica, 2024, 45(5): 508-516.
[15] WEN J H, WANG Z J.Short-term power load forecasting with hybrid TPA-BiLSTM prediction model based on CSSA[J]. Computer modeling in engineering & sciences, 2023, 136(1): 749-765.
[16] 张程, 林谷青, 黄靖, 等. 基于AMBOA-DBN结合相似日的短期光伏功率预测[J]. 太阳能学报, 2023, 44(6): 290-299.
ZHANG C, LIN G Q, HUANG J, et al.Short-term PV power prediction based on AMBOA-DBN combined with similar days[J]. Acta energiae solaris sinica, 2023, 44(6): 290-299.
[17] 唐宇峰, 陈星红, 蔡宇, 等. 基于时序分解和SSA-LSTM-Attention模型的尾矿坝位移预测[J]. 科学技术与工程, 2023, 23(29): 12753-12759.
TANG Y F, CHEN X H, CAI Y, et al.Tailings dam displacement prediction based on time-series decomposition and SSA-LSTM-Attention model[J]. Science technology and engineering, 2023, 23(29): 12753-12759.
[18] 周浩杰, 杨建卫, 王尊, 等. 基于LSTM光伏发电功率超短期预测模型研究[J]. 电源技术, 2023, 47(6): 785-789.
ZHOU H J, YANG J W, WANG Z, et al.Research on ultra-short term prediction model of photovoltaic power generation based on LSTM[J]. Chinese journal of power sources, 2023, 47(6): 785-789.
[19] CHEN X, DING K, ZHANG J W, et al.Online prediction of ultra-short-term photovoltaic power using chaotic characteristic analysis, improved PSO and KELM[J]. Energy, 2022, 248: 123574.
[20] 刘孝保, 严清秀, 易斌, 等. 基于集成学习和改进粒子群优化算法的流程制造工艺参数优化[J]. 中国机械工程, 2023, 34(23): 2842-2853.
LIU X B, YAN Q X, YI B, et al.Optimization of process parameters in process manufacturing based on ensemble learning and improved particle swarm optimization algorithm[J]. China mechanical engineering, 2023, 34(23): 2842-2853.
[21] KANG Z Z, DUAN R N, ZHENG Z M, et al.Grid aided combined heat and power generation system for rural village in North China plain using improved PSO algorithm[J]. Journal of cleaner production, 2024, 435: 140461.
[22] DOMÍNGUEZ R, VITALI S. Multi-chronological hierarchical clustering to solve capacity expansion problems with renewable sources[J]. Energy, 2021, 227: 120491.
[23] YANG M, ZHAO M, HUANG D W, et al.A composite framework for photovoltaic day-ahead power prediction based on dual clustering of dynamic time warping distance and deep autoencoder[J]. Renewable energy, 2022, 194: 659-673.
[24] 王晓霞, 俞敏, 冀明, 等. 基于气候相似性与SSA-CNN-LSTM的光伏功率组合预测[J]. 太阳能学报, 2023, 44(6): 275-283.
WANG X X, YU M, JI M, et al.Photovoltaic power combination forecasting based on climate similarity and SSA-CNN-LSTM[J]. Acta energiae solaris sinica, 2023, 44(6): 275-283.