CAJ | 학술논문

基于光波导边界传输特性和人工神经网络，文中介绍了一种可同时监测输电线路绝缘子表面等值盐密和灰密的污秽在线监测装置。在详细分析光波导表面污层对全反射光束影响机理的基础上，提出了光波导和光源设计方案。通过不同湿度、不同污秽等级、不同灰盐比条件下的积污标定试验，获取了大量样本数据，并依此建立和训练了人工神经网络模型。检验结果和运行实践表明，利用该装置及训练后的内嵌神经网络模型，可实现盐密和灰密的同时监测，且测试精度满足相关标准要求。
기우광파도변계전수특성화인공신경망락，문중개소료일충가동시감측수전선로절연자표면등치염밀화회밀적오예재선감측장치。재상세분석광파도표면오층대전반사광속영향궤리적기출상，제출료광파도화광원설계방안。통과불동습도、불동오예등급、불동회염비조건하적적오표정시험，획취료대량양본수거，병의차건립화훈련료인공신경망락모형。검험결과화운행실천표명，이용해장치급훈련후적내감신경망락모형，가실현염밀화회밀적동시감측，차측시정도만족상관표준요구。
An online pollution monitoring equipment is developed based on boundary transmission characteristics of optical waveguide and artificial neural network (ANN),which can measure both equivalent salt deposit density (ESDD) and non-soluble deposit density(NSDD)on insulators”surfaces.By analyzing the effects of the optical waveguide”s surface pollution on the reflected beam,the optical waveguide and the optical source are designed.A series of validating and calibrating tests have been made under different humidity,different salt density and ash density conditions,from which a large quantity of sample data are obtained and a unique ANN model is built and trained.The inspecting and running results show that this equipment together with the embedded ANN model can be used to monitor ESDD and NSDD of the insulators at once,and with satisfactory test precision.