摘要
针对电力负荷呈现出非线性的特性所导致预测精度不高等问题,本文提出基于VMD-QPSO-BiLSTM的短期电力负荷预测方法。首先,采用变分模态分解(VMD)降低负荷序列的非平稳性和复杂度;其次,基于量子粒子群算法(QPSO)改进的双向长短期记忆网络(BiLSTM)的方法进行预测;最后输出分解结果。另外,进行对比实验测试,实验表明本文所提的模型相比其他智能算法模型可以取得更高的预测精度。
Aiming at the problem of low prediction accuracy caused by the nonlinear characteristics of power load, this paper proposes a short-term power load forecasting method based on VMD-QPSO-BiLSTM. First, use Variational Modal Decomposition(VMD) to reduce the non-stationarity and complexity of the load sequence;secondly, use the improved bidirectional long short-term memory network(BiLSTM) method based on Quantum Particle Swarm Optimization(QPSO) for prediction;finally output decomposition result. In addition, a comparative experiment test is performed, and the experiment shows that the model proposed in this article can achieve higher prediction accuracy than other intelligent algorithm models.
作者
张铭飞
白苏赫
何艺萌
彭丹阳
李丁
ZHANG Mingfei;BAI Suhe;HE Yimeng;PENG Danyang;LI Ding(North China University of Water Resources and Electric Power,Zhengzhou Henan 450045,China)
出处
《信息与电脑》
2021年第19期47-49,共3页
Information & Computer
关键词
短期电力负荷预测
变分模态分解
量子粒子群算法
双向长短期记忆网络
short-term power load forecasting
variational modal decomposition
quantum particle swarm optimization
two-way long and short-term memory network
作者简介
张铭飞(1995-),男,湖南常德人,硕士研究生。研究方向:电力系统分析与控制。
