Photovoltaic Output Prediction Based on Improved Long Short-Term Memory Network Using White Shark Optimization Algorithm

doi:10.12096/j.2096-4528.pgt.24028

Abstract

Abstract:

Objectives In order to ensure the safety, stability, and economic operation of the power system after photovoltaic integration, a photovoltaic power prediction model based on improved long short-term memory (LSTM) network using white shark optimization algorithm is proposed, and uses the white shark optimization algorithm (WSO) to optimize the prediction model parameters. Methods To fully consider the environmental variables that affect photovoltaic output power, this study first uses the variational mode decomposition (VMD) method to decompose the environmental factor sequence, in order to reduce the non stationarity of the sequence. Next, the kernel principal component analysis (KPCA) method is used to extract the feature sequences of the main influencing factors and obtainthe optimal meteorological feature sequence. Finally, a LSTM network is used to predict multivariate feature sequences, and WSO is used to optimize the parameters of the LSTM to achieve accurate prediction of photovoltaic output. Results Compared with traditional photovoltaic power prediction methods, this model significantly improves the prediction accuracy of photovoltaic power generation. Conclusions Using WSO to optimize the number of hidden units, maximum training period, and initial learning rate in LSTM can effectively improve the efficiency and accuracy of optimization, providing a reference for the stable operation of high proportion photovoltaic systems after integration.

Key words: solar energy, photovoltaic power generation, photovoltaic output prediction, kernel principal component analysis, variational mode decomposition, long short-term memory network, white shark optimization algorithm

CLC Number:

TK 51

Chaoyang YAN, Lanqing LI, Haojia XU, Suo ZHUANG, Zhenhua ZHANG, Zirui RONG. Photovoltaic Output Prediction Based on Improved Long Short-Term Memory Network Using White Shark Optimization Algorithm[J]. Power Generation Technology, 2025, 46(4): 778-787.

Figures/Tables 10

Fig. 1 Photovoltaic output prediction process based on VMD-KPCA-WSO-LSTM

Tab. 1 Center frequencies corresponding to different m values

m	中心频率/kHz
2	8.8×10^-0.5、0.026 8
3	8.1×10^-0.5、0.025 9、0.054 8
4	8.1×10^-0.5、0.025 9、0.054 8、0.371 0
5	7.9×10^-0.5、0.025 9、0.053 4、0.084 1、0.399 1
6	7.9×10^-0.5、0.025 9、0.053 3、0.082 8、0.142 6、0.399 5
7	7.9×10^-0.5、0.025 9、0.053 3、0.082 5、0.137 1、0.210 2、0.420 3
8	7.9×10^-0.5、0.025 9、0.053 3、0.082 4、0.135 4、0.200 8 0.288 2、 0.422 1
9	7.9×10^-0.5、0.025 9、0.053 3、0.082 4、0.135 4、0.199 8、0.288 2、0.394 1、0.422 1

Fig. 2 The results of radiometric data decomposition

Fig. 3 The contribution rate of each feature after KPCA

Fig. 4 Comparison chart of the number of iterations

Fig. 5 Comparison chart of different prediction models on a sunny day

Tab. 2 Prediction error evaluation indicators for each model under sunny days

对比模型	$ε R M S E$	$ε M A E$	$ε M A P E$
LSTM	2.65	2.34	14.31
VMD-LSTM	2.37	2.15	12.99
VMD-KPCA-LSTM	2.78	2.50	14.38
VMD-KPCA-WSO-LSTM	0.82	0.70	3.78

Tab. 2 Prediction error evaluation indicators for each model under sunny days

对比模型	$ε R M S E$	$ε M A E$	$ε M A P E$
LSTM	2.65	2.34	14.31
VMD-LSTM	2.37	2.15	12.99
VMD-KPCA-LSTM	2.78	2.50	14.38
VMD-KPCA-WSO-LSTM	0.82	0.70	3.78

Fig. 6 Comparison chart of different prediction models in multi-cloud conditions

Fig. 7 Comparison of prediction errors of different prediction models in multi-cloud conditions

Tab. 3 Evaluation indicators of prediction error of each model in multi-cloud mode

对比模型	$ε R M S E$	$ε M A E$	$ε M A P E$
LSTM	7.04	5.40	54.53
VMD-LSTM	3.91	2.72	26.70
VMD-KPCA-LSTM	3.35	2.31	21.57
VMD-KPCA-WSO-LSTM	2.28	1.71	17.70

Tab. 3 Evaluation indicators of prediction error of each model in multi-cloud mode

对比模型	$ε R M S E$	$ε M A E$	$ε M A P E$
LSTM	7.04	5.40	54.53
VMD-LSTM	3.91	2.72	26.70
VMD-KPCA-LSTM	3.35	2.31	21.57
VMD-KPCA-WSO-LSTM	2.28	1.71	17.70

References 31

[1]	殷樾．基于粒子群算法-最小二乘支持向量机的日前光伏功率预测[J]．分布式能源，2021，6(2)：68-74．
	YIN Y ．Day-ahead photovoltaic power forecasting based on particle swarm optimization and least squares support vector machine[J]．Distributed Energy，2021，6(2)：68-74．
[2]	董明，李晓枫，杨章，等．基于数据驱动的分布式光伏发电功率预测方法研究进展[J]．电网与清洁能源，2024，40(1)：8-17．
	DONG M， LI X F， YANG Z，et al ．Research progress on data-driven prediction methods for distributed photovoltaic power generation[J]．Power System and Clean Energy，2024，40(1)：8-17．
[3]	张晓珂，张辉，戴小然，等．基于层次聚类和BILSTM的光伏短期功率预测模型[J]．智慧电力，2024，52(9)：41-48．
	ZHANG X K， ZHANG H， DAI X R，et al ．Photovoltaic short-term power forecasting model based on hierarchical clustering & BILSTM[J]．Smart Power，2024，52(9)：41-48．
[4]	关钦月，郑旭，徐敬友，等．规模新能源接入下考虑安稳风险评估的受端电网运行优化模型[J]．可再生能源，2023，41(7)：957-963．
	GUAN Q Y， ZHENG X， XU J Y，et al ．An operation optimization model of receiving power grid considering security risk assessment under large-scale new energy access[J]．Renewable Energy Resources，2023，41(7)：957-963．
[5]	马晓磊，李永光，段鹏飞，等．光伏发电参与电网供电功率稳定性调节技术研究[J]．电网与清洁能源，2023，39(11)：105-110．
	MA X L， LI Y G， DUAN P F，et al ．Research on the technology of photovoltaic power generation participating in power supply stability regulation of power grids[J]．Power System and Clean Energy，2023，39(11)：105-110．
[6]	陈凡，李智，丁津津，等．考虑光伏机理与数据驱动结合的短期功率预测[J]．科学技术与工程，2023，23(20)：8686-8692．
	CHEN F， LI Z， DING J J，et al ．Consider short-term power prediction combining photovoltaic mechanism and data-driven[J]．Science Technology and Engineering，2023，23(20)：8686-8692．
[7]	LIU Z， LIU X， ZHANG H，et al ．Integrated physical approach to assessing urban-scale building photovoltaic potential at high spatiotemporal resolution[J]．Journal of Cleaner Production，2023，388：135979． doi:10.1016/j.jclepro.2023.135979
[8]	王本涛，白杨，邢红涛，等．基于STL与MMoE多任务学习的区域多光伏电站超短期功率联合预测方法[J]．电力系统及其自动化学报，2022，34(9)：17-23．
	WANG B T， BAI Y， XING H T，et al ．Combined ultra-short-term power prediction method for regional multi-photovoltaic power stations based on STL and MMoE multi-task learning[J]．Proceedings of the CSU-EPSA，2022，34(9)：17-23．
[9]	MOHAMAD R P N L， AKHTER M N， MEKHILEF S，et al ．Review on the application of photovoltaic forecasting using machine learning for very short-to long-term forecasting[J]．Sustainability，2023，15(4)：2942． doi:10.3390/su15042942
[10]	李永飞，张耀，林帆，等．基于气候特征分析及改进XGBoost算法的中长期光伏电站发电量预测方法[J]．电力系统保护与控制，2024，52(11)：84-92．
	LI Y F， ZHANG Y， LIN F，et al ．Medium-and long-term power generation forecast based on climate characterisation and an improved XGBoost algorithm for photovoltaic power plants[J]．Power System Protection and Control，2024，52(11)：84-92．
[11]	战文华，车建峰，王勃，等．基于网格化数值天气预报的区域光伏发电多输出功率预测方法[J]．中国电力，2024，57(3)：144-151．
	ZHAN W H， CHE J F， WANG B，et al ．A grid-based numerical weather prediction method for multi-output prediction of regional photovoltaic power[J]．Electric Power，2024，57(3)：144-151．
[12]	杨锡运，马雪，张洋，等．基于EMD与加权马尔可夫链QR法的风电功率区间预测[J]．太阳能学报，2020，41(2)：66-72．
	YANG X Y， MA X， ZHANG Y，et al ．Probabilistic intervals forecasting of wind power based on EMD weighted Markov chain QR method[J]．Acta Energiae Solaris Sinica，2020，41(2)：66-72．
[13]	何文华，丁贵立，韩威，等．综合考虑电网碳效益和用户满意度的需求响应激励策略优化模型研究[J]．供用电，2023，40(10)：95-105．
	HE W H， DING G L， HAN W，et al ．Optimization model research of demand response incentive strategy considering carbon benefits of power grid and user satisfaction[J]．Distribution & Utilization，2023，40(10)：95-105．
[14]	闫宇露，程瑜，陈熙．适应光伏高渗透接入的配电网拓扑及储荷资源协同规划[J]．广东电力，2024，37(12)：50-60．
	YAN Y Y， CHENG Y， CHEN X ．Distribution network topology and coordinated planning of storage and load resources adapted to photovoltaic high penetration access[J]．Guangdong Electric Power，2024，37(12)：50-60．
[15]	林兴宇，肖迎群，张苏．基于极限学习机分位回归的光伏出力区间预测方法[J]．机械与电子，2023，41(6)：3-9．
	LIN X Y， XIAO Y Q， ZHANG S ．PV output interval forecasting method based on QRELM algorithm[J]．Machinery & Electronics，2023，41(6)：3-9．
[16]	张琰妮，史加荣，李津，等．融合残差与VMD-ELM-LSTM的短期风速预测[J]．太阳能学报，2023，44(9)：340-347．
	ZHANG Y N， SHI J R， LI J，et al ．Short-term wind speed prediction based on residual and VMD-ELM-LSTM[J]．Acta Energiae Solaris Sinica，2023，44(9)：340-347．
[17]	张恩辅，段冰冰，刘津平，等．基于改进麻雀搜索算法的优化型极限学习机[J]．软件工程，2023，26(9)：18-24．
	ZHANG E F， DUAN B B， LIU J P，et al ．An optimized extreme learning machine based on improved sparrow search algorithm[J]．Software Engineering，2023，26(9)：18-24．
[18]	胡丹，杨书恒．基于改进金枪鱼算法优化ELM模型的光伏功率预测[J]．武汉理工大学学报，2022，44(8)：97-104． doi:10.3963/j.issn.1671-4431.2022.08.015
	HU D， YANG S H ．Improved tuna algorithm to optimize ELM model for PV power prediction[J]．Journal of Wuhan University of Technology，2022，44(8)：97-104． doi:10.3963/j.issn.1671-4431.2022.08.015
[19]	LIANG L， SU T， GAO Y，et al ．FCDT-IWBOA-LSSVR：an innovative hybrid machine learning approach for efficient prediction of short-to-mid-term photovoltaic generation[J]．Journal of Cleaner Production，2023，385：135716． doi:10.1016/j.jclepro.2022.135716
[20]	邹晴，李乐，柳楠，等．基于混合卷积神经网络的多特征负荷预测方法研究[J]．电网与清洁能源，2024，40(9)：54-62．
	ZOU Q， LI L， LIU N，et al ．Research on the multi feature load forecasting method based on hybrid convolutional neural network[J]．Power System and Clean Energy，2024，40(9)：54-62．
[21]	曹煜祺，张立梅，白牧可．基于双层Elman神经网络的光伏发电功率预测[J]．供用电，2017，34(10)：8-13．
	CAO Y Q， ZHANG L M， BAI M K ．Photovoltaic power prediction based on double Elman neural network[J]．Distribution & Utilization，2017，34(10)：8-13．
[22]	俞娜燕，李向超，费科，等．基于SVR-UKF的光伏电站功率预测[J]．自动化与仪器仪表，2020(4)：73-77．
	YU N Y， LI X C， FEI K，et al ．Power prediction of photovoltaic power station based on support vector regression and unscented Kalman filter[J]．Automation & Instrumentation，2020(4)：73-77．
[23]	谢天宝，鲁云鹏，张颖茵．基于改进的杜鹃搜索算法优化支持向量机的10 kV并联电容器组故障诊断和预警研究[J]．自动化技术与应用，2019，38(4)：24-28．
	XIE T B， LU Y P， ZHANG Y Y ．Fault diagnosis and early warning study of 10 kV parallel capacitor bank based on improved cuckoo search algorithm optimizing SVM[J]．Techniques of Automation and Applications，2019，38(4)：24-28．
[24]	谢少华，何山，闫学勤，等．基于SSA-BP神经网络的光伏短期功率预测[J]．浙江工业大学学报，2022，50(6)：628-633．
	XIE S H， HE S， YAN X Q，et al ．Short term photovoltaic power prediction based on SSA-BP neural network[J]．Journal of Zhejiang University of Technology，2022，50(6)：628-633．
[25]	耿博，高贞彦，白恒远，等．结合相似日GA-BP神经网络的光伏发电预测[J]．电力系统及其自动化学报，2017，29(6)：118-123．
	GENG B， GAO Z Y， BAI H Y，et al ．PV generation forecasting combined with similar days and GA-BP neural network[J]．Proceedings of the CSU-EPSA，2017，29(6)：118-123．
[26]	田翠霞，黄敏，朱启兵．基于EMD-LMD-LSSVM联合模型的逐时太阳辐照度预测[J]．太阳能学报，2018，39(2)：504-512．
	TIAN C X， HUANG M， ZHU Q B ．Hourly solar irradiance forecast based on EMD-LMD-LSSVM joint model[J]．Acta Energiae Solaris Sinica，2018，39(2)：504-512．
[27]	余向阳，赵怡茗，杨宁宁，等．基于VMD-SE-LSSVM和迭代误差修正的光伏发电功率预测[J]．太阳能学报，2020，41(2)：310-318．
	YU X Y， ZHAO Y M， YANG N N，et al ．Photovoltaic power generation forecasting based on VMD-SE-LSSVM and iterative error correction[J]．Acta Energiae Solaris Sinica，2020，41(2)：310-318．
[28]	刘长良，武英杰，甄成刚．基于变分模态分解和模糊C均值聚类的滚动轴承故障诊断[J]．中国电机工程学报，2015，35(13)：3358-3365．
	LIU C L， WU Y J， ZHEN C G ．Rolling bearing fault diagnosis based on variational mode decomposition and fuzzy C means clustering[J]．Proceedings of the CSEE，2015，35(13)：3358-3365．
[29]	苏治，傅晓媛．核主成分遗传算法与SVR选股模型改进[J]．统计研究，2013，30(5)：54-62．
	SU Z， FU X Y ．Kernel principal component genetic algorithm and improved SVR stock selection model[J]．Statistical Research，2013，30(5)：54-62．
[30]	刘新忠，丁鹏达，曾慧林，等．一种基于灰狼策略的大白鲨混合优化算法[J]．成都工业学院学报，2023，26(4)：60-66．
	LIU X Z， DING P D， ZENG H L，et al ．A white shark hybrid optimization algorithm based on the grey wolf strategy[J]．Journal of Chengdu Technological University，2023，26(4)：60-66．
[31]	王鑫，吴际，刘超，等．基于LSTM循环神经网络的故障时间序列预测[J]．北京航空航天大学学报，2018，44(4)：772-784． doi:10.13700/j.bh.1001-5965.2017.0285
	WANG X， WU J， LIU C，et al ．Exploring LSTM based recurrent neural network for failure time series prediction[J]．Journal of Beijing University of Aeronautics and Astronautics，2018，44(4)：772-784． doi:10.13700/j.bh.1001-5965.2017.0285