

浏览全部资源
扫码关注微信
1.国网经济技术研究院有限公司,北京市 102209
2.国家电网有限公司,北京市 100031
3.新能源电力系统全国重点实验室(华北电力大学),北京市 102206
Received:09 October 2024,
Revised:2025-02-07,
Published Online:17 March 2025,
Published:10 August 2025
移动端阅览
刘栋,郭国栋,辛蜀骏等.基于KAN的可解释净负荷概率预测方法[J].电力系统自动化,2025,49(15):123-132.
LIU Dong,GUO Guodong,XIN Shujun,et al.Probabilistic Prediction Method for Interpretable Net Load Based on Kolmogorov-Arnold Network[J].Automation of Electric Power Systems,2025,49(15):123-132.
刘栋,郭国栋,辛蜀骏等.基于KAN的可解释净负荷概率预测方法[J].电力系统自动化,2025,49(15):123-132. DOI: 10.7500/AEPS20241009007.
LIU Dong,GUO Guodong,XIN Shujun,et al.Probabilistic Prediction Method for Interpretable Net Load Based on Kolmogorov-Arnold Network[J].Automation of Electric Power Systems,2025,49(15):123-132. DOI: 10.7500/AEPS20241009007.
分布式光伏在用户侧的大量接入给电力系统运行带来了强不确定性,增加了净负荷预测的难度。针对现有深度学习模型在净负荷概率预测时模型性能和解释能力难以权衡的问题,提出了一种基于Kolmogorov-Arnold网络(KAN)模型的净负荷概率预测方法。首先,通过对KAN模型的基函数权重矩阵和样条函数可学习系数的可解释性进行挖掘,实现了KAN模型的内在可解释性。然后,将最大似然估计损失函数作为模型设计依据,分别对KAN模型的均值和方差网络结构进行设计,实现对概率密度函数的有效拟合。同时,为量化多种气象特性对净负荷出力的影响,采用最大互信息系数方法进行特征筛选,增强了KAN模型对多种气象特征的映射能力。通过算例分析,验证了所提方法在净负荷功率概率预测上的有效性和可靠性,且相较于传统方法,所提方法在模型准确性、可解释能力、训练效率以及泛化能力上均具备更好的性能。
The large-scale integration of distributed photovoltaics on the user side has brought strong uncertainties to the operation of the power system
increasing the difficulty of net load prediction. Aiming at the problem that it is difficult to balance the model performance and interpretability in the net load probabilistic prediction of existing deep learning models
this paper proposes a probabilistic prediction method for net loads based on the Kolmogorov-Arnold network (KAN) model. Firstly
by exploring the interpretability of the basis function weight matrix and the learnable coefficients of the spline function of the KAN model
the inherent interpretability of the KAN model is achieved. Then
taking the maximum likelihood estimation loss function as the basis for model design
the mean and variance network structures of the KAN model are designed respectively to achieve an effective fit of the probability density function. Meanwhile
to quantify the impact of various meteorological characteristics on the net load output
the maximum mutual information coefficient method is adopted for feature screening
which enhances the mapping capability of the KAN model for various meteorological characteristics. Through case study
the effectiveness and reliability of the proposed method in net load power probabilistic prediction are verified. Moreover
compared with traditional methods
the proposed method has better performance in terms of model accuracy
interpretability
training efficiency
and generalization capability.
钱平凡 , 周健奇 . 整县推进屋顶分布式光伏开发模式亟须调整 [EB/OL].[ 2024-10-05 ]. https://www.drc.gov.cn/DocView.aspx?chnid=379&leafid=1338&docid=2907225 https://www.drc.gov.cn/DocView.aspx?chnid=379&leafid=1338&docid=2907225 .
QIAN Pingfan , ZHOU Jianqi . The county-wide promotion of rooftop distributed photovoltaic development models needs urgent adjustment [EB/OL]. [ 2024-10-05 ]. https://www.drc.gov.cn/DocView.aspx?chnid=379&leafid=1338&docid=2907225 https://www.drc.gov.cn/DocView.aspx?chnid=379&leafid=1338&docid=2907225 .
王同光 , 秦海岩 . 2014—2023年中国风电产业发展成就及未来展望 [J]. 太阳能 , 2024 ( 7 ): 31 - 39 .
WANG Tongguang , QIN Haiyan . Development achievement and future prospect of wind power industry in China from 2014 to 2023 [J]. Solar Energy , 2024 ( 7 ): 31 - 39 .
LOPEZ L J , LIU X Q , MORROW D J . Spatial aggregation of small-scale photovoltaic generation using Voronoi decomposition [J]. IEEE Transactions on Sustainable Energy , 2020 , 11 ( 4 ): 2677 - 2686 .
ERDENER B C , FENG C , DOUBLEDAY K , et al . A review of behind-the-meter solar forecasting [J]. Renewable and Sustainable Energy Reviews , 2022 , 160 : 112224 .
WANG X Q , DU X , WANG H Y , et al . Research on coordinated optimization of source-load-storage considering renewable energy and load similarity [J]. Energies , 2024 , 17 ( 6 ): 1301 .
WANG Y , ZHANG N , CHEN Q X , et al . Data-driven probabilistic net load forecasting with high penetration of behind-the-meter PV [J]. IEEE Transactions on Power Systems , 2018 , 33 ( 3 ): 3255 - 3264 .
LI K P , YAN J C , HU L , et al . Two-stage decoupled estimation approach of aggregated baseline load under high penetration of behind-the-meter PV system [J]. IEEE Transactions on Smart Grid , 2021 , 12 ( 6 ): 4876 - 4885 .
孙元章 , 吴俊 , 李国杰 , 等 . 基于风速预测和随机规划的含风电场电力系统动态经济调度 [J]. 中国电机工程学报 , 2009 , 29 ( 4 ): 41 - 47 .
SUN Yuanzhang , WU Jun , LI Guojie , et al . Dynamic economic dispatch considering wind power penetration based on wind speed forecasting and stochastic programming [J]. Proceedings of the CSEE , 2009 , 29 ( 4 ): 41 - 47 .
TOUBEAU J F , BOTTIEAU J , VALLEE F , et al . Deep learning-based multivariate probabilistic forecasting for short-term scheduling in power markets [J]. IEEE Transactions on Power Systems , 2019 , 34 ( 2 ): 1203 - 1215 .
万灿 , 崔文康 , 宋永华 . 新能源电力系统概率预测:基本概念与数学原理 [J]. 中国电机工程学报 , 2021 , 41 ( 19 ): 6493 - 6509 .
WAN Can , CUI Wenkang , SONG Yonghua . Probabilistic forecasting for power systems with renewable energy sources: basic concepts and mathematical principles [J]. Proceedings of the CSEE , 2021 , 41 ( 19 ): 6493 - 6509 .
王瑞临 , 赵健 , 孙智卿 , 等 . 考虑数据分布偏移的短期居民净负荷预测方法 [J]. 电力建设 , 2024 , 45 ( 2 ): 90 - 101 .
WANG Ruilin , ZHAO Jian , SUN Zhiqing , et al . Short-term resident net load forecasting method considering data distribution deviation [J]. China Industrial Economics , 2024 , 45 ( 2 ): 90 - 101 .
杨茂 , 董昊 . 基于数值天气预报风速和蒙特卡洛法的短期风电功率区间预测 [J]. 电力系统自动化 , 2021 , 45 ( 5 ): 79 - 85 .
YANG Mao , DONG Hao . Short-term wind power interval prediction based on wind speed of numerical weather prediction and Monte Carlo method [J]. Automation of Electric Power Systems , 2021 , 45 ( 5 ): 79 - 85 .
万灿 , 宋永华 . 新能源电力系统概率预测理论与方法及其应用 [J]. 电力系统自动化 , 2021 , 45 ( 1 ): 2 - 16 .
WAN Can , SONG Yonghua . Theories, methodologies and applications of probabilistic forecasting for power systems with renewable energy sources [J]. Automation of Electric Power Systems , 2021 , 45 ( 1 ): 2 - 16 .
董骁翀 , 孙英云 , 蒲天骄 , 等 . 基于时序混合密度网络的超短期风电功率概率预测 [J]. 电力系统自动化 , 2022 , 46 ( 14 ): 93 - 100 .
DONG Xiaochong , SUN Yingyun , PU Tianjiao , et al . Ultra-short-term probabilistic forecasting of wind power based on temporal mixture density network [J]. Automation of Electric Power Systems , 2022 , 46 ( 14 ): 93 - 100 .
SANJARI M J , GOOI H B . Probabilistic forecast of PV power generation based on higher order Markov chain [J]. IEEE Transactions on Power Systems , 2017 , 32 ( 4 ): 2942 - 2952 .
YAN J , LI K , BAI E W , et al . Hybrid probabilistic wind power forecasting using temporally local Gaussian process [J]. IEEE Transactions on Sustainable Energy , 2016 , 7 ( 1 ): 87 - 95 .
GUAN C , LUH P B , MICHEL L D , et al . Hybrid Kalman filters for very short-term load forecasting and prediction interval estimation [J]. IEEE Transactions on Power Systems , 2013 , 28 ( 4 ): 3806 - 3817 .
丁藤 , 冯冬涵 , 林晓凡 , 等 . 基于修正后ARIMA-GARCH模型的超短期风速预测 [J]. 电网技术 , 2017 , 41 ( 6 ): 1808 - 1814 .
DING Teng , FENG Donghan , LIN Xiaofan , et al . Ultra-short-term wind speed forecasting based on improved ARIMA-GARCH model [J]. Power System Technology , 2017 , 41 ( 6 ): 1808 - 1814 .
SUN M C , FENG C , ZHANG J . Conditional aggregated probabilistic wind power forecasting based on spatio-temporal correlation [J]. Applied Energy , 2019 , 256 : 113842 .
SHUAI H , LI F X . Physics-informed Kolmogorov-Arnold networks for power system dynamics [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2408.06650v1 https://arxiv.org/abs/2408.06650v1 .
SALINAS D , FLUNKERT V , GASTHAUS J , et al . DeepAR: probabilistic forecasting with autoregressive recurrent networks [J]. International Journal of Forecasting , 2020 , 36 ( 3 ): 1181 - 1191 .
KEMKER R , MCCLURE M , ABITINO A , et al . Measuring catastrophic forgetting in neural networks [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/1708.02072v4 https://arxiv.org/abs/1708.02072v4 .
FAN F L , XIONG J , LI M , et al . On interpretability of artificial neural networks: a survey [J]. IEEE Transactions on Radiation and Plasma Medical Sciences , 2021 , 5 ( 6 ): 741 - 760 .
王小君 , 窦嘉铭 , 刘曌 , 等 . 可解释人工智能在电力系统中的应用综述与展望 [J]. 电力系统自动化 , 2024 , 48 ( 4 ): 169 - 191 .
WANG Xiaojun , DOU Jiaming , LIU Zhao , et al . Review and prospect of explainable artificial intelligence and its application in power systems [J]. Automation of Electric Power Systems , 2024 , 48 ( 4 ): 169 - 191 .
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 Xiangjing , ZHOU Wenxin , LI Chaojie , 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 .
HUANG B , WANG J H . Applications of physics-informed neural networks in power systems—a review [J]. IEEE Transactions on Power Systems , 2023 , 38 ( 1 ): 572 - 588 .
GAO Y , RUAN Y J . Interpretable deep learning model for building energy consumption prediction based on attention mechanism [J]. Energy and Buildings , 2021 , 252 : 111379 .
GARREAU D , VON LUXBURG U . Explaining the explainer: a first theoretical analysis of LIME [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2001.03447v2 https://arxiv.org/abs/2001.03447v2 .
LIU Z M , WANG Y X , VAIDYA S , et al . KAN: Kolmogorov-Arnold networks [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2404.19756v5 https://arxiv.org/abs/2404.19756v5 .
LIU Z M , MA P C , WANG Y X , et al . KAN 2 . 0 : Kolmogorov-Arnold networks meet science[EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2408.10205v1 https://arxiv.org/abs/2408.10205v1 .
HAN X , ZHANG X , WU Y , et al . KAN4TSF: are KAN and KAN-based models effective for time series forecasting? [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2408.11306?context=cs.LG https://arxiv.org/abs/2408.11306?context=cs.LG .
KICH V A , BOTTEGA J A , STEINMETZ R , et al . Kolmogorov-Arnold network for online reinforcement learning [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2408.04841v3 https://arxiv.org/abs/2408.04841v3 .
SHEN S , YOUNES R . Reimagining linear probing: Kolmogorov-Arnold networks in transfer learning [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2409.07763v1 https://arxiv.org/abs/2409.07763v1 .
YANG X Y , WANG X C . Kolmogorov-Arnold transformer [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2409.10594v1 https://arxiv.org/abs/2409.10594v1 .
ZHOU Z H , LI Y Y , GUO Z L , et al . A white-box deep-learning method for electrical energy system modeling based on Kolmogorov-Arnold network [EB/OL]. [ 2024-09-05 ]. https://arxiv.org/abs/2409.08044v1 https://arxiv.org/abs/2409.08044v1 .
MACHLEV R , HEISTRENE L , PERL M , et al . Explainable artificial intelligence (XAI) techniques for energy and power systems: review, challenges and opportunities [J]. Energy and AI , 2022 , 9 : 100169 .
GNEITING T , RAFTERY A E . Strictly proper scoring rules, prediction, and estimation [J]. Journal of the American Statistical Association , 2007 , 102 ( 477 ): 359 - 378 .
RESHEF D N , RESHEF Y A , FINUCANE H K , et al . Detecting novel associations in large data sets [J]. Science , 2011 , 334 ( 6062 ): 1518 - 1524 .
0
Views
0
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
京公网安备11010802024621