| 注册
首页|期刊导航|电网技术|基于概率预测与随机响应面法的新能源孤岛配电网实时风险评估与调控策略

基于概率预测与随机响应面法的新能源孤岛配电网实时风险评估与调控策略

梁远升 程康 王钢 李海锋 张思捷 徐真理 徐征

电网技术2023,Vol.47Issue(12):4948-4957,10.
电网技术2023,Vol.47Issue(12):4948-4957,10.DOI:10.13335/j.1000-3673.pst.2023.1294

基于概率预测与随机响应面法的新能源孤岛配电网实时风险评估与调控策略

Real-time Risk Assessment and Regulation Strategy of New Energy Islanded Distribution Network Based on Probabilistic Prediction and Stochastic Response Surface Methodology

梁远升 1程康 1王钢 1李海锋 1张思捷 1徐真理 1徐征1

作者信息

  • 1. 华南理工大学电力学院,广东省广州市 510641
  • 折叠

摘要

Abstract

The uncertainty of high proportion of distributed generation brings great challenges to the stable operation of the islanded distribution networks.In this paper,for the short-term prediction of source and load based on the traditional distribution model has the shortcomings of sharp peaks and heavy tails,the distribution model of source and load prediction error under multi-scenarios and different time scales is constructed by combining the bidirectional long and short-term memory(BiLSTM)neural network with the nonparametric kernel density method(KDE).On this basis,the system operates and regulates in multi-temporal operation and control process,taking into account the short-term meteorological fluctuations in the uncertainty.Then,the distflow model is relaxed by using the mixed integer second-order cone programming and the probabilistic Distflow of the system is obtained from the stochastic response surface(SRSM).Based on the stochastic response surface to improve the Sobol'method,the global sensitivity analysis model of the isolated island distribution network operation risk taking into account the source and load uncertainty is established.For this reason,a risk real-time risk assessment and regulation strategy based on BiLSTM-SRSM method is proposed.Finally,the feasibility of the proposed method is verified using a radial distribution system with the 1EEE33 nodes.

关键词

双向长短时记忆神经网络/实时风险预警/随机响应面法/非参数核密度/全局灵敏度

Key words

bidirectional long/short-term neural network/real-time risk warning/stochastic response surface method/kernel density estimation/global sensitivity

分类

信息技术与安全科学

引用本文复制引用

梁远升,程康,王钢,李海锋,张思捷,徐真理,徐征..基于概率预测与随机响应面法的新能源孤岛配电网实时风险评估与调控策略[J].电网技术,2023,47(12):4948-4957,10.

基金项目

广东省基础与应用基础研究基金项目(2023A1515011035).Project Supported by the Guangdong Basic and Applied Basic Research Foundation(2023A1515011035). (2023A1515011035)

电网技术

OA北大核心CSCDCSTPCD

1000-3673

访问量0
|
下载量0
段落导航相关论文