摘要
为了减少反时限过流保护整定值过于依赖人工经验,缓解保护动作之间配合程度较低、电流稳定性较差的问题,提出基于自适应神经元的配电网输电线路反时限过流保护优化整定方法。分析输电线路运行模式和线路故障存在的不确定性,计算发生故障的概率;通过探究反向荷载率、电源容量、谐波含量等指标确定输电线的实际承载力;构建反时限过流保护动作方程,确定反时限过流保护的3种特征;以主保护与后备保护动作时差最小为目标,构建目标函数,确立约束条件;设计具有三个神经元的神经网络模型,每个神经元对应一种保护特征,引入转换模块,将输入量变换为状态量,通过神经元学习,输出最终的过流保护优化整定结果。比较实验结果表明:应用所提方法,随着故障电流增大,保护动作时间曲线未出现相交状况;在测试刚开始时,电流波形出现波动,但迅速恢复到平稳。所提方法能够使保护动作之间有效配合,输电线电流可以始终保持在稳定状态,为保护定值优化提供了新的研究思路。
In order to reduce the excessive dependence of anti-time overcurrent protection setting value on manual experience,alleviate the problems of low cooperation between protection actions and poor current stability,an optimal set method for anti-time overcurrent protection of distribution network transmission lines based on adaptive neurons is proposed.This paper analyzes the uncertainty of transmission line operation mode and line failure,and calculates the probability of failure.The actual bearing capacity of the transmission line is determined by exploring the reverse load rate,power supply capacity and harmonic content.The action equation of anti-time overcurrent protection is constructed to determine the three characteristics of anti-time overcurrent protection.With the goal of minimizing the time difference between the main protection and backup protection actions,an objective function is constructed and the constraint conditions are established.A neural network model is constructed with three neurons,each neuron corresponds to a protection feature,a transformation module is introduced to transform the input quantity into a state quantity,and output the final overcurrent protection optimal setting result through neuron learning.The comparative experiment results show that under the application of the proposed method,with the increase of fault current,the curves of protection action time do not intersect.The current waveform fluctuates at the beginning of the test,but quickly returns to plateau.The proposed method can make the protection actions cooperate effectively,and the transmission line current can always be kept in a stable state,which provides a new research idea for the optimization of the protection fixed value.
作者
孙刚
孙迪康
SUN Gang;SUN Dikang(Tancheng County Power Supply Company of State Grid Shandong Electric Power Company,Tancheng 276100,China;School of Electrical&Information Engineering,Northeast Petroleum University,Daqing 163319,China)
出处
《微型电脑应用》
2024年第7期226-229,233,共5页
Microcomputer Applications
关键词
自适应神经元
配电网输电线路
反时限过流保护
谐波含量
adaptive neuron
distribution network transmission line
anti-time overcurrent protection
harmonic content
