Residential photovoltaic power generation prediction model based on PSO-BP neural network

doi:10.3969/j.issn.2097-0706.2025.06.009

Abstract

Abstract:

The residential rooftop，as an idle space that is almost free from shading，provides ideal conditions for the deployment of photovoltaic （PV） systems. However，the intermittency and volatility of photovoltaic generation，along with the mismatch between photovoltaic power generation and residential electricity demand at different times，present significant challenges for the energy management system in achieving supply-demand balance. As an essential component of energy system optimization and performance enhancement，photovoltaic generation forecasting is critical to effectively addressing these challenges. To this end，this paper proposes an improved forecasting model that combines Particle Swarm Optimization （PSO） with Backpropagation （BP） Neural Networks. In this model，PSO is employed to optimize the parameters of the BP neural network，significantly improving the accuracy and stability of photovoltaic power prediction. Experimental results demonstrate that the improved model outperforms the traditional BP neural network in forecasting accuracy across all seasons. The average root mean square error （RMSE） is reduced by 42.31%，and the coefficient of determination （R²） increases by 2.22%. The annual average forecasting accuracy exceeds 90.00%，with the highest accuracy achieved in winter，reaching 99.46%. This study provides reliable forecasting data for the optimized scheduling of photovoltaic systems in residential buildings，offering substantial practical application value.

Key words: photovoltaic power generation, power prediction, BP Neural Network, PV residential building, four-season prediction model

CLC Number:

TK019

BAN Fengchun, CHEN Xiaofeng, HUANG Zhijia. Residential photovoltaic power generation prediction model based on PSO-BP neural network[J]. Integrated Intelligent Energy, 2025, 47(6): 85-93.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Table 1

Fig.8

Table 2

Table 3

Fig.9

Fig.10

Fig.11

Fig.12

Table 4

Comparison of BP and PSO-BP prediction results

季节	评价指标	预测结果		提高率/%
季节	评价指标	BP	PSO-BP	提高率/%
春	$R M S E$	2.05	1.03	-49.62
春	R²	0.971 7	0.996 8	2.59
夏	$R M S E$	1.45	1.06	-26.90
夏	R²	0.988 03	0.994 09	0.61
秋	$R M S E$	2.98	1.22	-59.06
秋	R²	0.942 45	0.989 23	4.96
冬	$R M S E$	1.07	0.71	-33.64
冬	R²	0.987 49	0.994 59	0.72

Table 4

References 34

[1]	劳凯月, 祁明亮, 高敏刚, 等. 碳中和目标下我国光伏发电发展路径规划研究[J]. 科技促进发展, 2023, 19(4): 219-227.
	LAO Kaiyue, QI Mingliang, GAO Mingang, et al. Study on China's optimal solar photovoltaic power development path to achieve carbon neutrality[J]. Science and Technology for Development, 2023, 19(4): 219-227.
[2]	张春明, 晋家润, 武春旭, 等. 能源转型视角下垭口村“零碳乡村” 可行性研究[J]. 建筑科学, 2025, 41(2): 34-45.
	ZHANG Chunming, JIN Jiarun, WU Chunxu, et al. Feasibility study on the "zero-carbon village" of yakou village from the perspective of energy transition[J]. Building Science, 2025, 41(2): 34-45.
[3]	WEN Q M, LIU G, RAO Z H, et al. Applications, evaluations and supportive strategies of distributed energy systems: A review[J]. Energy and Buildings, 2020, 225: 110314.
[4]	NADEEM TBIN, SIDDIQUI M, KHALID M, et al. Distributed energy systems: A review of classification, technologies, applications, and policies[J]. Energy Strategy Reviews, 2023, 48: 101096.
[5]	曾鸣, 许文秀, 魏阳, 等. 基于可再生能源分布式发电投资的模糊综合评价研究[J]. 华东电力, 2011, 39(2): 180-183.
	ZENG Ming, XU Wenxiu, WEI Yang, et al. Research on fuzzy synthetic evaluation of renewable distributed energy generation investment[J]. East China Electric Power, 2011, 39(2): 180-183.
[6]	李莎, 鲍丹阳. 负载自适应扰动孤岛检测法[J]. 电力电子技术, 2025, 59(1): 86-89.
	LI Sha, BAO Danyang. Load-adaptive disturbance islanding detection method[J]. Power Electronics, 2025, 59(1): 86-89.
[7]	SOMMERFELD J, BUYS L, VINE D. Residential consumers' experiences in the adoption and use of solar PV[J]. Energy Policy, 2017, 105: 10-16.
[8]	CHI F A, WU Y. Residential buildings integrated with SPVP and RCL towards energy saving[J]. Renewable and Sustainable Energy Reviews, 2025, 212: 115370.
[9]	黄志甲, 寇遵丽, 郝红婷. 民居能源中心能量匹配及其鲁棒性[J]. 太阳能学报, 2021, 42(11): 495-500.
	HUANG Zhijia, KOU Zunli, HAO Hongting. Energy matching and robustness of home energy hub[J]. Acta Energiae Solaris Sinica, 2021, 42(11): 495-500.
[10]	NELEGA R, GREU D I, JECAN E, et al. Prediction of power generation of a photovoltaic power plant based on neural networks[J]. IEEE Access, 2023, 11: 20713-20724.
[11]	TONG J L, XIE L P, FANG S X, et al. Hourly solar irradiance forecasting based on encoder-decoder model using series decomposition and dynamic error compensation[J]. Energy Conversion and Management, 2022, 270: 116049.
[12]	张永宁, 任晓颖, 张飞, 等. 基于深度学习的光伏功率预测技术[J/OL]. 综合智慧能源, 1-8 (2024-04-01)[2025-02-12]. https://kns.cnki.net/kcms/detail/41.1461.TK.20240328.2244.002.html.
	ZHANG Yongning, REN Xiaoying, ZHANG Fei, et al. Deep learning based Photovoltaic power prediction technology[J/OL]. Integrated Intelligent Energy, 1-8 (2024-04-01)[2025-02-12]. https://kns.cnki.net/kcms/detail/41.1461.TK.20240328.2244.002.html.
[13]	袁俊球, 王迪, 谢小锋, 等. 基于广义天气分类的ICEEMDAN-LSTM网络光伏发电功率短期预测[J]. 综合智慧能源, 2024, 46(9): 53-60.
	YUAN Junqiu, WANG Di, XIE Xiaofeng, et al. Study of short-term PV power prediction based on ICEEMDAN-LSTM networks under generalized weather classifications[J]. Integrated Intelligent Energy, 2024, 46(9): 53-60.
[14]	OGLIARI E, DOLARA A, MANZOLINI G, et al. Physical and hybrid methods comparison for the day ahead PV output power forecast[J]. Renewable Energy, 2017, 113: 11-21.
[15]	MASSERAN N, RAZALI A M, IBRAHIM K, et al. Fitting a mixture of von mises distributions in order to model data on wind direction in peninsular malaysia[J]. Energy Conversion and Management, 2013, 72: 94-102.
[16]	IMPRAM S, VARBAK NESE S, ORAL B. Challenges of renewable energy penetration on power system flexibility: A survey[J]. Energy Strategy Reviews, 2020, 31: 100539.
[17]	MANSOUR AAIT, TILIOUA A, TOUZANI M. Bi-LSTM, GRU and 1D-CNN models for short-term photovoltaic panel efficiency forecasting case amorphous silicon grid-connected PV system[J]. Results in Engineering, 2024, 21: 101886.
[18]	YANG C, LI S Y, GOU Z H. Spatiotemporal prediction of urban building rooftop photovoltaic potential based on GCN-LSTM[J]. Energy and Buildings, 2025, 334: 115522.
[19]	黄牧涛, 邢芳菲, 陈兴邦, 等. 基于K-means聚类和极限学习机组合算法的短期光伏功率预测[J]. 水电能源科学, 2024, 42(2): 217-220, 216.
	HUANG Mutao, XING Fangfei, CHEN Xingbang, et al. Short-term PV power prediction based on K-means clustering and extreme learning machine combination algorithm[J]. Water Resources and Power, 2024, 42(2): 217-220, 216.
[20]	NGUYEN-DUC T, DO-DINH H, FUJITA G, et al. Multi 2D-CNN-based model for short-term PV power forecast embedded with Laplacian Attention[J]. Energy Reports, 2024, 12: 2086-2096.
[21]	DEWI T, MARDIYATI E N, RISMA P, et al. Hybrid machine learning models for pv output prediction: Harnessing random forest and LSTM-RNN for sustainable energy management in aquaponic system[J]. Energy Conversion and Management, 2025, 330: 119663.
[22]	IHEANETU K J. Solar photovoltaic power forecasting: A review[J]. Sustainability, 2022, 14(24): 17005.
[23]	KHELIFI R, GUERMOUI M, RABEHI A, et al. Short-term PV power forecasting using a hybrid TVF-EMD-ELM strategy[J]. International Transactions on Electrical Energy Systems, 2023, 2023(1): 6413716.
[24]	HU K Y, WANG L D, LI W J, et al. Forecasting of solar radiation in photovoltaic power station based on ground-based cloud images and BP neural network[J]. IET Generation, Transmission and Distribution, 2022, 16(2): 333-350.
[25]	LIU Z F, LIU H B, ZHANG D M. PSO-BP-based optimal allocation model for complementary generation capacity of the photovoltaic power station[J]. Energy Engineering, 2023, 120(7): 1717-1727.
[26]	LI G H, WEI X, YANG H. Decomposition integration and error correction method for photovoltaic power forecasting[J]. Measurement, 2023, 208: 112462.
[27]	程先龙, 张杰, 李思莹, 等. 基于VMD-BP-BiLSTM的短期风电功率预测[J/OL]. 综合智慧能源, 1-10 (2024-12-12)[2025-02-12]. https://kns.cnki.net/kcms/detail/41.1461.TK.20241211.1634.002.html.
	CHENG Xianlong, ZHANG Jie, LI Siying, et al. Short-term wind power prediction based on VMD-BP-BiLSTM[J/OL]. Integrated Intelligent Energy, 1-10 (2024-12-12)[2025-02-12]. https://kns.cnki.net/kcms/detail/41.1461.TK.20241211.1634.002.html.
[28]	颜高洋, 丁贵立, 许志浩, 等. 基于帝王蝶优化算法的BP神经网络能源预测模型研究[J]. 南昌工程学院学报, 2023, 42(3): 88-94.
	YAN Gaoyang, DING Guili, XU Zhihao, et al. Research on BP neural network energy prediction model based on monarch butterfly optimization algorithm[J]. Journal of Nanchang Institute of Technology, 2023, 42(3): 88-94.
[29]	王登海, 安玥馨, 廖晨博, 等. 基于CNN-LSTM混合神经网络的光伏发电量预测方法研究[J]. 西安石油大学学报(自然科学版), 2024, 39(1): 129-134.
	WANG Denghai, AN Yuexin, LIAO Chenbo, et al. Research on photovoltaic power generation prediction method based on CNN-LSTM hybrid neural network[J]. Journal of Xi'an Shiyou University (Natural Science Edition), 2024, 39(1): 129-134.
[30]	袁冰清, 程功, 郑柳刚. BP神经网络基本原理[J]. 数字通信世界, 2018(8): 49-51.
	YUAN Bingqing, CHENG Gong, ZHENG Liugang. Basic principle of BP neural networks[J]. Digital Communication World, 2018(8): 49-51.
[31]	胡祖源, 靳现林, 谭雅之, 等. 基于改进粒子群算法的分布式光伏及储能系统优化配置[J]. 综合智慧能源, 2023, 45(1): 49-57.
	HU Zuyuan, JIN Xianlin, TAN Yazhi, et al. Optimized configuration of distributed photovoltaic and energy storage system based on improved particle swarm algorithm[J]. Integrated Intelligent Energy, 2023, 45(1): 49-57.
[32]	中国气象局. 气候季节划分: QX/T 152—2012[S]. 北京: 气象出版社, 2012.
[33]	GB/T 9535-2005/IEC 61215:2005 地面用晶体硅光伏组件—设计鉴定和定型[S]. 北京: 中国标准出版社, 2005.
[34]	陈美珍, 柳扬, 徐胜彬, 等. 基于天气类型及改进的小波神经网络的光伏发电短期预测[J]. 贵州大学学报(自然科学版), 2022, 39(1): 70-77, 88.
	CHEN Meizhen, LIU Yang, XU Shengbin, et al. Short term forecasting of photovoltaic power generation based on weather type and improved wavelet neural network[J]. Journal of Guizhou University (Natural Sciences), 2022, 39(1): 70-77, 88.

参数	数值
太阳电池种类	单晶硅
太阳电池组件尺寸/（mm×mm×mm）	1 762×1 134×30
峰值功率/W	430
工作电压/V	31.63
工作电流/A	13.60
开路电压/V	38.14
短路电流/A	14.32
组件效率/%	21.52

算法	参数	数值
BP	训练次数	1 000
	学习速率	0.1
	最小误差	0.000 01
PSO	C₁	2
	C₂	1.8
	ω	0.9
	迭代次数	50
	种群规模	10
	粒子最大速度	0.1
	粒子最小速度	-0.1
	权重和偏置范围	[-1,1]

季节	温度范围/℃	季节总时段	训练集时间范围	测试集时间范围
春	10 ~25	3月1日— 6月14日	3月1日— 6月12日	6月13日— 6月14日
夏	≥25	6月15日— 8月31日	6月15日— 8月29日	8月30日— 8月31日
秋	10 ~25	9月1日— 11月30日	9月1日— 11月28日	11月29日— 11月30日
冬	≤10	12月1日—次年2月28/29日	12月1日—次年2月26/27日	2月27/28日— 2月28/29日