摘要
为保证输电线路的安全可靠运行,提出一种基于场强修正模型的输电线路弧垂监测技术。首先针对场强测量仪性能易受外界环境影响的问题,提出一种基于超限学习机的输电线路下地面场强数据修正技术,该技术利用超限学习机训练一个单隐层前馈神经网络,用于数据修正,以解决场强测量仪测量数据不理想的问题。其次,在三相输电导线三维电场计算模型下,基于输电导线下方测量点的实时电场数据,采用模拟退火算法反演得到输电线路的弧垂值。再次,为确保及时准确地发现输电线路故障,构建弧垂电力监测系统,通过四大模块协同工作,对弧垂状况进行分析、监测和预警。最后,对所提出的修正和反演技术进行仿真,实验结果表明,所提出的技术可将反演误差降至2.21%,在有效排除气候因素的同时提高了弧垂计算精度,进而保证了电力系统的安全可靠性。
In order to ensure the safe and reliable operation of transmission lines,a sag monitoring technology based on the electric field correction model is proposed.This technique firstly takes the case into considerations that the external environmental factors influence the electric field measurement data greatly.A single-hidden layer feedforward neural network (SLFN)is trained based on the extreme learning machine to correct the data measured by the field strength measuring instrument.Secondly,based on the three-dimensional electric field calculation model of three-phase transmission conductor,according to the real-time measurement data of the measuring point under the transmission wire,the sag value of the transmission line is obtained by using the simulated annealing (SA)algorithm.Thirdly,an electric sag monitoring system is built for analyzing the running situation of the sag to guarantee the normal work of transmission lines.Finally,experiments are executed to verify the validity of the algorithm and the results show that the proposed technique can reduce the inversion error to 2.21%,which effectively eliminates the climatic factors and improves the sag calculation accuracy,thus ensuring the safety and reliability of the power system.
作者
许阳
赵彬
夏阳
郝新宇
李从林
Xu Yang;Zhao Bin;Xia Yang;Hao Xinyu;Li Conglin(State Grid Shaanxi Electric Power Company Xi'an Power Supply Company,Xi'an 710032,China;State Grid Shaanxi Electric Power Company Shangluo Power Supply Company,Shangluo 726000,China)
出处
《电子测量技术》
2018年第21期75-80,共6页
Electronic Measurement Technology
关键词
输电线路
模拟退火
弧垂
超限学习机
前馈神经网络
transmission line
simulated annealing (SA)
sag
extreme learning machine(ELM)
single-hidden layer feedforward neural network(SLFN)
