摘要
由于配电网短路故障产生的运维成本逐年增加,影响电力企业整体经济效益,为此提出机器学习下配电网短路故障快速智能恢复方法。分析回声状态网络,建立配电网短路故障恢复模型;设置配电网运行指标约束条件,构建回声状态网络目标函数,计算函数中个体与约束条件间的违背程度,得到实际目标函数;在目标优化阶段,保证重要负荷最大化,再考虑配电网运行均衡度,以此实现配电网短路故障快速智能恢复。对所提方法展开实验测试,结果表明,所提方法可有效恢复故障支路和节点,同时恢复后电压、电流波动均保证在所允许的范围内。
As the operation and maintenance cost of distribution network short-circuit fault increases year by year,which affects the overall economic benefits of power enterprises,a fast intelligent recovery method of distribution network short-circuit fault based on machine learning is proposed.The echo state network is analyzed,and the short-circuit fault recovery model of distribution network is established.The distribution network operation index constraint conditions are set,the echo state network objective function is constructed,and the violation degree between the individual and the constraint conditions in the function is calculated to obtain the actual objective function.In the target optimization stage,the important load is maximized,and then the operation balance of the distribution network is considered,so as to realize the quick and intelligent recovery of the short-circuit fault of the distribution network.The experimental results of the proposed method show that the proposed method can effectively restore the fault branches and nodes,and the voltage and current fluctuations are guaranteed within the allowable range after restoration.
作者
任庭昊
REN Ting-hao(Zunyi Power Supply Bureau of Guizhou Power Grid Co.,Ltd.,Zunyi 563000 China)
出处
《自动化技术与应用》
2025年第1期45-48,75,共5页
Techniques of Automation and Applications
基金
贵州电网有限责任公司科技创新项目(060300KK52170009)。
关键词
机器学习
配电网短路故障
智能恢复
目标函数
目标优化
machine learning
short circuit fault of power grid
Smart Recovery
objective function
target optimization
作者简介
任庭昊(1982-),男,本科,工程师,研究方向:电力系统调度自动化和配网自动化。
