一种新型日前光伏功率概率区间预测方法研究  

Research on a New Photovoltaic Power Probability Interval Prediction Method

作　　者：李溪 胡梦浩 冯耀冬 LI Xi;HU Menghao;FENG Yaodong(School of Electrical Engineering,North China University of Water Resource and Electric Power,Zhengzhou,Henan 450011,China)

机构地区：[1]华北水利水电大学电气工程学院,河南郑州450011

出　　处：《自动化应用》2023年第3期71-74,78,共5页Automation Application

摘　　要：面对大量光伏出力并网,其带来的波动性和不确定性给智能微电网的控制和设计带来了很大的挑战。因此,文章提出一种基于WD-BiLSTM-NPKDE的日前光伏功率概率区间预测方法。首先,利用小波分解方法进行数据信号分解,进一步消除随机性、波动性对预测精度的影响;其次,将分解后的分量输入到BiLSTM模型进行训练;再次,采用非参数核密度估计求解概率预测结果;最后,得到给定置信水平下的光伏功率预测区间。算例结果表明,该方法能有效得到光伏功率区间。In the face of a large number of photovoltaic output grid-connected,the volatility and uncertainty brought by it bring great challenges to the control and design of smart micro-grids.Therefore,a photovoltaic power probability interval prediction method based on WD-BiLSTM-NPKDE is proposed.Firstly,the wavelet decomposition method is used to decompose the data signal to further eliminate the influence of randomness and volatility on the prediction accuracy.Secondly,the decomposed components are input to the BiLSTM model for training.Then,nonparametric kernel density estimation is used to solve the probability prediction results.Finally,the PV power prediction interval at the given confidence level is obtained.The results show that the proposed method can effectively obtain the photovoltaic power range.

关 键 词：光伏功率概率区间预测 小波分解 BiLSTM模型 非参数核密度估计 

分 类 号：TM615[电气工程—电力系统及自动化]