基于GEP滑动窗口模型的变压器油中溶解气体含量预测  

Prediction of gases content dissolved in power transformer oil based on GEP sliding window model

作　　者：胡资斌[1,2] 朱永利[1] 段振锋[2] 董卓[3] 

机构地区：[1]华北电力大学电气与电子工程学院,河北保定071003 [2]湖南省电力公司娄底电业局,湖南娄底417000 [3]湖南省电力公司超高压管理局,湖南长沙410000

出　　处：《华北电力大学学报（自然科学版）》2012年第4期42-46,共5页Journal of North China Electric Power University：Natural Science Edition

摘　　要：为了对变压器的运行状态和潜伏性故障进行有效预测,提出了基于基因表达式程序设计(Gene Expression Programming,GEP)滑动窗口预测模型的变压器油中溶解气体浓度的预测方法。根据变压器油中溶解气体量的变化特点,选择合适的嵌入维度、终结点集、函数集等GEP运行参数后由适应度函数驱动进行遗传操作,演化出各气体的预测模型。结合变压器运行实例,给出了7种主要气体的预测结果以及H2含量的预测公式,并与MGM(1,7)模型进行比较。对比结果表明,该方法能有效地提高预测精度。In order to predicting the operational status and the latent faults of a power transformer effectively, a new method to forecast the dissolved gas＇ s concentration in transformer oil based on GEP sliding window model is proposed. According to the change characteristics of the dissolved gases＇ concentration in transformer oil, appropriate embedding dimension, terminals, functions and other running parameters of GEP are selected, then gas forecasting models driven by the fitness function for genetic operation can be evolved. A real power transformer is taken as an example, and prediction results for seven major gases and the prediction formula of H2 are given in this paper. Contrasts with the MGM （ 1, 7） model are given and comparison show that GEP model can improve the prediction accuracy effectively.

关 键 词：GEP 滑动窗口 油中溶解气体 浓度预测 

分 类 号：TM411[电气工程—电器]