摘要
随着大规模以光伏、风电为主的分布式能源接入和负荷需求的复杂化,传统合环电流计算方法难以满足精度要求和反映合环特性的不确定性,本文提出了一种基于残差修正的合环电流预测方法。针对大量分布式能源接入和负荷需求复杂化的影响,采用自适应噪声完全集合经验模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise,CEEMDAN)对数据进行处理,分解为稳定分量以提取更多内部特征信息。构建CNN-BiLSTM神经网络模型,利用卷积神经网络(Convolutional Neural Network,CNN)提取数据全局特征,通过双向长短期记忆神经网络(Bidirectional Long Short-Term Memory,BiLSTM)捕捉数据长期依赖关系,进行合环电流时段预测。最后,为减小由于源荷的不确定性和出力波动性导致的误差,引入马尔科夫链模型对预测结果进行残差修正。仿真结果表明,该方法能够有效提高合环电流预测精度,证明所提方法的优越性和可行性。
With the complexity of large-scale distributed energy access and load demand dominated by photovoltaic and wind power,it is difficult for the traditional combined loop current calculation method to meet the accuracy requirements and reflect the uncertainty of combined loop characteristics.A combined loop current prediction method based on residual correction was proposed.In order to address the impact of large-scale distributed energy sources and complex load demand,the data were processed using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN),which decomposes them into stable components to extract more internal feature information.The CNN-BiLSTM neural network model was constructed to extract the global features of the data using Convolutional Neural Network(CNN),and to capture the long-term data dependencies through Bidirectional Long Short-Term Memory(BiLSTM),realizing the combined loop current time period prediction.Finally,in order to reduce the errors due to the uncertainty of source load and output volatility,a Markov chain model was introduced to correct the residuals of the prediction results.Simulation results showed that the method can effectively improve the prediction accuracy of the combined loop current,proving the superiority and feasibility of the proposed method.
作者
龙欣雨
舒征宇
陈昌松
姚钦
童华敏
LONG Xinyu;SHU Zhengyu;CHEN Changsong;YAO Qin;TONG Huamin(College of Electrical Engineering and New Energy,China Three Gorges University,Yichang 443002,China;Yichang Power Supply Company,State Gri d Hubei Electric Power Co.,Ltd.,Yi chang 443000,China)
出处
《国外电子测量技术》
2025年第2期17-24,共8页
Foreign Electronic Measurement Technology
