电力系统自动化2025，Vol.49Issue(19)：75-85,11.DOI:10.7500/AEPS20250213005
融合模糊神经网络预决策的有源配电网实时随机调度方法
Real-time Stochastic Dispatch Method for Active Distribution Network Using Fuzzy Neural Network Pre-decision
摘要
Abstract
With the large-scale integration of distributed resources into the active distribution network,internal uncertainties of the active distribution network increase.To establish an accurate dispatch model,the uncertainty of photovoltaic output and the randomness of load are described as the uncertainty of corresponding prediction errors,and the probability distribution of uncertain variables is obtained through data-driven methods.Considering that the power flow model based on second-order cone relaxation(SOCR)may violate relaxation constraints and introduce errors,the power flow model is re-derived based on the Euler equations,thereby further establishing a stochastic optimal power dispatch model for active distribution networks that is both economical and secure.Aiming at the characteristics of the proposed model,this paper presents a real-time stochastic dispatch method for active distribution networks using fuzzy neural network(FNN)pre-decision.First,the FNN is employed to provide a fuzzy description of the probability distribution of uncertain variables,and its output is used as the initial value for solver optimization.Then,the solver is utilized to accelerate the solution process.Finally,the effectiveness of the proposed model and method is validated through a modified IEEE 33-bus system.关键词
有源配电网/实时随机调度/数据驱动/潮流/模糊神经网络/不确定性Key words
active distribution network/real-time stochastic dispatch/data-driven/power flow/fuzzy neural network/uncertainty引用本文复制引用
程礼临,罗子杰,李群,张宁宇,李雅然,臧海祥..融合模糊神经网络预决策的有源配电网实时随机调度方法[J].电力系统自动化,2025,49(19):75-85,11.基金项目
国家电网公司科技项目(4000-202418061A-1-1-ZN). This work is supported by State Grid Corporation of China(No.4000-202418061A-1-1-ZN). （4000-202418061A-1-1-ZN）
