半月刊

ISSN 1000-1026

CN 32-1180/TP

+高级检索 English
新能源电网中考虑特征选择的Bi-LSTM网络短期负荷预测
作者:
作者单位:

安徽省新能源利用与节能省级实验室(合肥工业大学),安徽省 合肥市 230009

摘要:

新能源电网中负荷对各特征因素更为敏感,当面对海量特征数据时,短期负荷预测方法面临着新的挑战。针对含有高维特征数据的新能源电网,提出一种考虑特征选择的双向长短期记忆(Bi-LSTM)网络短期负荷预测方法。先将样本数据按密度进行聚类后映射到权重诱导空间中,通过定义一种间隔量度数据结构,以间隔之和最大为目标函数。为实现解空间的稀疏性,将正则项添加到目标函数中,并采用梯度下降法求解特征权值。经过预试验确定特征选择阈值等超参数,从而选出所需的特征因素。最后,使用Bi-LSTM网络基于选择后的数据进行负荷预测。以中国某地区新能源电网为例,验证了该方法的有效性,结果表明其与传统方法相比,具有更好的准确性和适用性。

关键词:

基金项目:

国家重点研发计划资助项目(2016YFB0901100);国家自然科学基金区域创新发展联合基金资助项目(U19A20106)。

通信作者:

作者简介:


Short-term Load Forecasting in Renewable Energy Grid Based on Bi-directional Long-short-term Memory Network Considering Feature Selection
Author:
Affiliation:

(Anhui Provincial Laboratory of Renewable Energy Utilization and Energy Saving (Hefei University of Technology), Hefei 230009, China)

Abstract:

Loads in the renewable energy grid is more sensitive to various feature factors, it is a new challenge for short-term load forecasting method when facing the massive feature data. Aiming at the in the renewable energy grid with high-dimensional feature data, a short-term load forecasting method based on bi-directional long-short-term memory(Bi-LSTM) network considering feature selection is proposed. The sample data are firstly clustered and mapped into the weight-induced space according to density, by defining a structure of interval measurement data, the maximum interval sum is used as the objective function. In order to achieve the sparsity of the solution space, a regular term is added into the objective function, and the feature weights are solved by the gradient descent algorithm. Hyper-parameters such as the feature selection threshold are determined by pre-tests, and then the required feature factors are selected. Finally, the Bi-LSTM network is used to carry out load forecasting based on the selected data. Taking the renewable energy grid in a certain area of China as an example, the effectiveness of the proposed method is verified, compared with the traditional methods, the proposed method is more accurate and applicative.

Keywords:

Foundation:
This work is supported by the National Key R&D Program of China (No. 2016YFB0901100) and the Regional Innovation and Development Joint Fund of National Natural Science Foundation of China (No. U19A20106).
引用本文
[1]杨龙,吴红斌,丁明,等.新能源电网中考虑特征选择的Bi-LSTM网络短期负荷预测[J/OL].电力系统自动化,http://doi. org/10.7500/AEPS20200202002.
YANG Long, WU Hongbin, DING Ming, et al. Short-term Load Forecasting in Renewable Energy Grid Based on Bi-directional Long-short-term Memory Network Considering Feature Selection[J/OL]. Automation of Electric Power Systems, http://doi. org/10.7500/AEPS20200202002.
复制
分享
历史
  • 收稿日期:2020-02-02
  • 最后修改日期:2020-06-30
  • 录用日期:2020-04-23
  • 在线发布日期:
  • 出版日期:
相关附件