基于模糊神经网络的大型电力变压器风险评估方法研究.docx

第3 9卷 第3 9卷 第5期 湖南大学学报(自然科学版) V01．39，No．5 2 O l 2 年5月 Journal of Hunan University(Natural Sciences) May．2 O 1 2 文章编号：1674—2974(2012)05·0059一06 基于模糊神经网络的大型电力变压器风险 评估方法研究’ 俞乾1甜，李卫国1，罗日成3 (1．华北电力大学高电压与电磁兼容北京市重点实验室，北京 102206； 2．湖南省 电力公司永州电业局，湖南永州 425000；3．长沙理工大学电气与信息工程学院，湖南长沙410014) 接要：为实时、客观地衡量在役变压器故障所带来的风险，从评估的准确性和快捷性的 角度，提出了基于模糊层次分析法和人工神经网络相结合的变压器风险评估方法．通过分析影 响变压器风险水平的风险因素，构建了变压器风险评估的层次分析模型，在模型中建立风险评 估指标体系，并结合模糊一致判断矩阵，定量计算出变压器风险分析中各风险因素的权重；利 用人工神经网络的自适应能力与非线性映射能力，将权重大的风险因素作为神经网络的输入， 实现变压器风险的智能化量化评估．仿真结果表明，该方法提高了风险评估的速度和结果的准 确度，能为变压器的风险管理与维修决策提供可行的决策依据． 关键词：模糊层次分析法；人工神经网络；风险评估；电力变压器 中图分类号：TM407 文献标识码：A A Risk Assessment Method of Power Transformer Based on Fuzzy Analytic Hierarchy Process and Neural Network YU Qianl·2t，LI Wei—gu01，LUO Ri—chen93 (1．蹦iIlg Key Laboratory of H曲v01tage＆EMc．№ml曲如Electric Po州。U山，Beij堍 102206．mm； 2．Hunan E1ectric Company Yongzhou Electric Power Bureau．Yor培zhou。Hunan 425000。China； 3．Sch001 of E1ectrical＆Informtion EngineerilIg。Changsha Unjv of Science and Technology．Changsha-Hu吐n 410014。China) Abstr扯t：For the real—time and objective measurement of the risk caused by on—line power transfomer， power transformer risk assessment method was put forward based Fuzzy AnaIytic Hierarchy Process (FAHP)and Artificial Neural Network(ANN)from the perspective of the accuracy and quickness of risk assessment． An analytic hierarchy process model for transfomer risk assessment was buiIt by anaIyzing risk factors affecting transfomer risk level． In this model，a risk index system was established and index weights were quantified． The weight relation of each risk factor in transfomer risk calculation was lyzed by applying fuzzy consistency judgment matrix．1t took major risk factors the input of the neutral network by making of adaptive ability and nonlinear mapping ability of the ANN，thus realizing the in— telIigent quantitative assessment of tmnsformer risks． Simulation results have shown that this method in— the speed and result accuracy of ri