
浏览全部资源
扫码关注微信
南京理工大学自动化学院,江苏省南京市 210094
权浩(1985—),男,博士,教授,博士生导师,主要研究方向:不确定性量化、预测、可再生能源集成、人工智能在电力系统中的应用。E-mail:quanhao@njust.edu.cn
吕立臻(1993—),男,硕士研究生,主要研究方向:预测、人工智能和机器学习。E-mail:lvlizhen112@njust.edu.cn
郭健(1974—),男,通信作者,博士,教授,博士生导师,主要研究方向:智能控制与智能系统、机器人系统、高精度电机控制。E-mail:guoj1002@njust.edu.cn
收稿日期:2021-09-01,
修回日期:2022-03-23,
网络出版日期:2022-08-17,
纸质出版日期:2022-10-25
移动端阅览
权浩,吕立臻,郭健等.基于自然进化策略的可再生能源ELM预测两阶段优化训练方法[J].电力系统自动化,2022,46(20):83-91.
QUAN Hao,LYU Lizhen,GUO Jian,et al.Two-stage Optimal Training Method of Extreme Learning Machine Forecasting Based on Natural Evolution Strategy for Renewable Energy[J].Automation of Electric Power Systems,2022,46(20):83-91.
权浩,吕立臻,郭健等.基于自然进化策略的可再生能源ELM预测两阶段优化训练方法[J].电力系统自动化,2022,46(20):83-91. DOI: 10.7500/AEPS20210901010.
QUAN Hao,LYU Lizhen,GUO Jian,et al.Two-stage Optimal Training Method of Extreme Learning Machine Forecasting Based on Natural Evolution Strategy for Renewable Energy[J].Automation of Electric Power Systems,2022,46(20):83-91. DOI: 10.7500/AEPS20210901010.
随着中国“双碳”目标的提出和可再生能源接入比例的不断提高,其精确预测对提高新能源消纳水平和电网稳定性以及降低系统规划运行成本具有重要意义。为此,提出了基于自然进化策略(NES)的极限学习机(ELM)网络参数两阶段优化训练方法及可再生能源功率预测改进模型。在第1阶段,使用ELM随机参数和广义矩阵逆运算过程训练模型网络参数。ELM训练过程具有随机性且矩阵逆运算过程缺少对参数的微调,导致模型虽然训练速度较快,但在精度上有所欠缺。因此,第2阶段在原有模型训练结果的基础上引入输出层偏移和自适应的NES算法,对网络参数进行二次优化调整。结果表明
该两阶段优化算法能够有效地提高模型预测精度。
With the proposal “carbon emission peak and carbon neutrallity” target of China and the continuous increase of the proportion of renewable energy access
accurate forecasting is of great significance to improve the level of renewable energy accommodation and grid stability
and reduce the system planning and operation cost. Therefore
a two-stage optimal training method of extreme learning machine (ELM) network parameters based on natural evolution strategy (NES) and an improved model of renewable energy power forecasting are proposed. In the first stage
random parameters and the generalized matrix inverse operation process of ELM are used to train the model network parameters. Due to the randomness of the ELM training process and the lack of fine-tuning for parameters in the matrix inverse operation process
the accuracy is deficient although the training speed of the model is fast. Therefore
in the second stage
based on the training results of the original model
the output layer bias and adaptive NES algorithm are introduced to optimize and adjust the network parameters for the second time. The results show that the proposed two-stage optimization algorithm can effectively improve the accuracy of the forecasting model
田智宇 . “十四五”推进绿色发展的挑战、机遇与路径 [J]. 中国经贸导刊(中) , 2020 ( 12 ): 106 - 108 .
TIAN Zhiyu . Challenges, opportunities and paths for promoting green development in the “14th Five-Year Plan ”[J]. China Economic & Trade Herald , 2020 ( 12 ): 106 - 108 .
康重庆 , 姚良忠 . 高比例可再生能源电力系统的关键科学问题与理论研究框架 [J]. 电力系统自动化 , 2017 , 41 ( 9 ): 2 - 11 .
KANG Chongqing , YAO Liangzhong . Key scientific issues and theoretical research framework for power systems with high proportion of renewable energy [J]. Automation of Electric Power Systems , 2017 , 41 ( 9 ): 2 - 11 .
QUAN H , SRINIVASAN D , KHOSRAVI A . Short-term load and wind power forecasting using neural network-based prediction intervals [J]. IEEE Transactions on Neural Networks and Learning Systems , 2014 , 25 ( 2 ): 303 - 315 .
YANG D Z , JIRUTITIJAROEN P , WALSH W M . Hourly solar irradiance time series forecasting using cloud cover index [J]. Solar Energy , 2012 , 86 ( 12 ): 3531 - 3543 .
DONG Z B , YANG D Z , REINDL T , et al . A novel hybrid approach based on self-organizing maps, support vector regression and particle swarm optimization to forecast solar irradiance [J]. Energy , 2015 , 82 : 570 - 577 .
周念成 , 廖建权 , 王强钢 , 等 . 深度学习在智能电网中的应用现状分析与展望 [J]. 电力系统自动化 , 2019 , 43 ( 4 ): 180 - 191 .
ZHOU Niancheng , LIAO Jianquan , WANG Qianggang , et al . Analysis and prospect of deep learning application in smart grid [J]. Automation of Electric Power Systems , 2019 , 43 ( 4 ): 180 - 191 .
HUANG G B , ZHU Q Y , SIEW C K . Extreme learning machine: theory and applications [J]. Neurocomputing , 2006 , 70 ( 1/2/3 ): 489 - 501 .
HORNIK K . Approximation capabilities of multilayer feedforward networks [J]. Neural Networks , 1991 , 4 ( 2 ): 251 - 257 .
HORNIK K , STINCHCOMBE M , WHITE H . Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks [J]. Neural Networks , 1990 , 3 ( 5 ): 551 - 560 .
LESHNO M , LIN V Y , PINKUS A , et al . Multilayer feedforward networks with a nonpolynomial activation function can approximate any function [J]. Neural Networks , 1993 , 6 ( 6 ): 861 - 867 .
HUANG G B . Learning capability and storage capacity of two-hidden-layer feedforward networks [J]. IEEE Transactions on Neural Networks , 2003 , 14 ( 2 ): 274 - 281 .
TAMURA S , TATEISHI M . Capabilities of a four-layered feedforward neural network: four layers versus three [J]. IEEE Transactions on Neural Networks , 1997 , 8 ( 2 ): 251 - 255 .
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 .
叶林 , 马明顺 , 靳晶新 , 等 . 考虑风电-光伏功率相关性的因子分析-极限学习机聚合方法 [J]. 电力系统自动化 , 2021 , 45 ( 23 ): 31 - 40 .
YE Lin , MA Mingshun , JIN Jingxin , et al . Factor analysis-extreme learning machine aggregation method considering correlation of wind power and photovoltaic power [J]. Automation of Electric Power Systems , 2021 , 45 ( 23 ): 31 - 40 .
文云峰 , 赵荣臻 , 肖友强 , 等 . 基于多层极限学习机的电力系统频率安全评估方法 [J]. 电力系统自动化 , 2019 , 43 ( 1 ): 133 - 140 .
WEN Yunfeng , ZHAO Rongzhen , XIAO Youqiang , et al . Frequency safety assessment of power system based on multi-layer extreme learning machine [J]. Automation of Electric Power Systems , 2019 , 43 ( 1 ): 133 - 140 .
陆俊 , 陈志敏 , 龚钢军 , 等 . 基于极限学习机的居民用电行为分类分析方法 [J]. 电力系统自动化 , 2019 , 43 ( 2 ): 97 - 104 .
LU Jun , CHEN Zhimin , GONG Gangjun , et al . Classification analysis method for electricity consumption behavior based on extreme learning machine algorithm [J]. Automation of Electric Power Systems , 2019 , 43 ( 2 ): 97 - 104 .
杨锡运 , 关文渊 , 刘玉奇 , 等 . 基于粒子群优化的核极限学习机模型的风电功率区间预测方法 [J]. 中国电机工程学报 , 2015 , 35 ( 增刊1 ): 146 - 153 .
YANG Xiyun , GUAN Wenyuan , LIU Yuqi , et al . Prediction intervals forecasts of wind power based on PSO-KELM [J]. Proceedings of the CSEE , 2015 , 35 ( Supplement 1 ): 146 - 153 .
曾亮 , 雷舒敏 , 王珊珊 , 等 . 基于OVMD-SSA-DELM-GM模型的超短期风电功率预测方法 [J]. 电网技术 , 2021 , 45 ( 12 ): 4701 - 4712 .
ZENG Liang , LEI Shumin , WANG Shanshan , et al . Ultra-short-term wind power prediction based on OVMD-SSA-DELM-GM model [J]. Power System Technology , 2021 , 45 ( 12 ): 4701 - 4712 .
WIERSTRA D , SCHAUL T , PETERS J , et al . Natural evolution strategies [C]// IEEE Congress on Evolutionary Computation , June 1-6 , 2008 , Hong Kong, China : 3381 - 3387 .
WIERSTRA D , SCHAUL T , GLASMACHERS T , et al . Natural evolution strategies [J]. The Journal of Machine Learning Research , 2014 , 15 ( 1 ): 949 - 980 .
SCHYMURA C , KOLOSSA D . Learning dynamic stream weights for linear dynamical systems using natural evolution strategies [C]// IEEE International Conference on Acoustics, Speech and Signal Processing , May 12-17 , 2019 , Brighton, UK : 7893 - 7897 .
MASUTOMI K , NAGATA Y , ONO I . Extending distance-weighted exponential natural evolution strategy for function optimization in uncertain environments [C]// IEEE Congress on Evolutionary Computation , June 20-23 , 2013 , Cancun, Mexico : 2122 - 2129 .
KOBAYASHI Y , ONO I . Sequential estimation of states and parameters of nonlinear state space models using particle filter and natural evolution strategy [C]// IEEE Congress on Evolutionary Computation , July 19-24 , 2020 , Glasgow, UK : 1 - 8 .
0
浏览量
0
下载量
3
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621