摘要
根据各溶解气体的在线监测数据,采用灰色GM(1,1)和PNN融合技术进行在线故障诊断.先通过GM(1,1)预测模型预测未来时刻变压器中矿物绝缘油在电和热的作用下,分解产生的氢、甲烷、乙烷、乙烯及乙炔5种气体溶解浓度,并将预测结果作为概率神经网络故障诊断的输入利用PNN进行变压器故障诊断.实例表明,该方法能够诊断变压器在未来时刻的绝缘状况,并能满足工程实际需要.
Based on the online inspection data of various soluble gases, an integrated technology of GM （ 1,1 ） and PNN is introduce into the online fault diagnosis. Firstly, GM （ 1,1 ） forecasts the concentration of H2, CH4, C2H6, C2 H2 and C2 H4 , which are produced by decomposing the mineral insnlative oil under the effect of electricity and heat. At the same time, the forecasting results are chosen as the input data for the fault diagnosis through the PNN. The actual example shows that the method can be used to diagnose the insulation condition of transformer in the future time and can also meet the real oroiect reauirements.
出处
《嘉兴学院学报》
2008年第6期99-104,共6页
Journal of Jiaxing University
