以油中气体为特征量的变压器绝缘故障的模糊诊断方法及应用研究
|
摘要
变压器作为电力系统的枢纽设备，其运行可靠性直接关系到电力系统的安全与稳定。本文针对应用油中溶解气体分析方法进行变压器绝缘监督时所遇到的主要技术难点，提出了提高变压器故障诊断的准确性、可靠性的几种模诊断方法，主要研究内容有：
通过对判断变压器故障常用的三比值法和改良电协研法的深入分析，其诊断准确率较高，但对故障原因、故障现象和故障机理间同时存在不确定性和模糊性的变压器等电气设备的故障诊断，难于满足工程应用的需要；在处理变压器等结构复杂设备的多症状、多原因故障诊断时，模糊关系矩阵可以全面反映这种症状与故障类型间的因果关系，进而提出了变压器故障诊断的模糊综合诊断模型；同时作者还深入分析了模糊算子的特性；针对常用的模糊评判结果的模糊集的集化方法的不足性，提出了将模糊综合诊断与模糊规则推理结合起来进行故障诊断的方法，能达到较好的效果。即首先采用模糊推理对故障原因进行“过滤”，滤掉可能性极小的原因，然后再利用规则推理对剩下的可能原因进行验证，得出最终结论。
根据以变压器DGA数据为特征量的样本空间各样本差异特性以及样本在空间R~s的分布特性，首次提出了基于势函数自适应加权的变压器绝缘故障诊断的模糊c-均值聚类模型；同时，从s维样本空间的F~c-划分几何特性出发，提出了一种求取样本集的类势有效邻域半径和自适应求取聚类数和聚类中心初值的方法；对一个待诊断样本，设计了基于类势密度函数意义下的属性测度和诊断准则。
经过大量的诊断实例表明：本文的模糊综合诊断与模糊规则推理结合起来的故障诊断方法对电力研究变压器绝缘故障诊断是有效的，能大大减少错判、误判比率；自适应加权模糊聚类方法能对变压器的故障样本正确分类，同时能有效判别故障类型，诊断的准确性大大高于常用的三比值法。
The operation reliability of the power transformer,which is the major equipment in power system,directly related to the safety and stability of whole power system. In accordance with the technological difficulties encountered in the process of insulation supervision based on the Dissolved Gases Analysis (DGA),several kinds of model and method are presented to improve the reliability and precision of fault diagnosis of the power transformer. Main research content includes:
By deeply studying the common transformer faults diagnosing methods,such as three-ratio methods and improved electrical committee agreements,several shortcomings such as uncertainness judgment when the fault reasons,phenomenon and principles come out together while can not consistent to each other etc. For this reason,the old methods can not fully meet the need to engineering practical application.
Considering fuzzy relationship matrix can fully represents the causality between fault symptoms and fault types,when diagnosing complex equipments with multiple symptoms and fault causes such as power transformer,a synthetic fuzzy diagnosing model is firstly proposed to diagnose transformer's insulation faults based on DGA in this paper. Meanwhile,by further study the character of fuzzy factors,aimed at correcting the defects in classifying fuzzy sets of common fuzzy judgment,a method are put forward by unifying fuzzy synthetic diagnosis and fuzzy principle reasoning. Firstly,fuzzy principle reasoning is applied to filter fault reasons with low possibility,then fuzzy synthetic diagnosis is utilized to test all the left fault reasons and pick up the reasonable ones.
In connection with the difference and distribution characteristic of the samples in sample space RS based on DGA,a new self-adapted weight fuzzy omean clustering model of fault diagnosis of the power transformer based on the potential function is proposed. Meanwhile,from the aspect of geometry characteristic of FC-divided in s dimension sample space,a method is proposed for the purpose of getting an effective adjacent radius,adaptive cluster number c and original cluster center of X sample set. For the diagnosis
sample x,the property measure and diagnosis rule are proposed,which under the condition of potential density function that determine c number of optimal fuzzy cluster P1.
Plenty of diagnosis examples demonstrate the validity of the gray relationship analysis method in diagnosing power system insulation faults. Not only the misjudgment rate can be reduced,but also the transformer fault incidences can be classified by using self-adapted weight fuzzy clustering method. Further more,the transformer insulation condition can be analyzed,as well as the fault position can be located by recognizing the fault types. The diagnosing precision is much higher comparing with former three-ratio method.
通过对判断变压器故障常用的三比值法和改良电协研法的深入分析，其诊断准确率较高，但对故障原因、故障现象和故障机理间同时存在不确定性和模糊性的变压器等电气设备的故障诊断，难于满足工程应用的需要；在处理变压器等结构复杂设备的多症状、多原因故障诊断时，模糊关系矩阵可以全面反映这种症状与故障类型间的因果关系，进而提出了变压器故障诊断的模糊综合诊断模型；同时作者还深入分析了模糊算子的特性；针对常用的模糊评判结果的模糊集的集化方法的不足性，提出了将模糊综合诊断与模糊规则推理结合起来进行故障诊断的方法，能达到较好的效果。即首先采用模糊推理对故障原因进行“过滤”，滤掉可能性极小的原因，然后再利用规则推理对剩下的可能原因进行验证，得出最终结论。
根据以变压器DGA数据为特征量的样本空间各样本差异特性以及样本在空间R~s的分布特性，首次提出了基于势函数自适应加权的变压器绝缘故障诊断的模糊c-均值聚类模型；同时，从s维样本空间的F~c-划分几何特性出发，提出了一种求取样本集的类势有效邻域半径和自适应求取聚类数和聚类中心初值的方法；对一个待诊断样本，设计了基于类势密度函数意义下的属性测度和诊断准则。
经过大量的诊断实例表明：本文的模糊综合诊断与模糊规则推理结合起来的故障诊断方法对电力研究变压器绝缘故障诊断是有效的，能大大减少错判、误判比率；自适应加权模糊聚类方法能对变压器的故障样本正确分类，同时能有效判别故障类型，诊断的准确性大大高于常用的三比值法。
The operation reliability of the power transformer,which is the major equipment in power system,directly related to the safety and stability of whole power system. In accordance with the technological difficulties encountered in the process of insulation supervision based on the Dissolved Gases Analysis (DGA),several kinds of model and method are presented to improve the reliability and precision of fault diagnosis of the power transformer. Main research content includes:
By deeply studying the common transformer faults diagnosing methods,such as three-ratio methods and improved electrical committee agreements,several shortcomings such as uncertainness judgment when the fault reasons,phenomenon and principles come out together while can not consistent to each other etc. For this reason,the old methods can not fully meet the need to engineering practical application.
Considering fuzzy relationship matrix can fully represents the causality between fault symptoms and fault types,when diagnosing complex equipments with multiple symptoms and fault causes such as power transformer,a synthetic fuzzy diagnosing model is firstly proposed to diagnose transformer's insulation faults based on DGA in this paper. Meanwhile,by further study the character of fuzzy factors,aimed at correcting the defects in classifying fuzzy sets of common fuzzy judgment,a method are put forward by unifying fuzzy synthetic diagnosis and fuzzy principle reasoning. Firstly,fuzzy principle reasoning is applied to filter fault reasons with low possibility,then fuzzy synthetic diagnosis is utilized to test all the left fault reasons and pick up the reasonable ones.
In connection with the difference and distribution characteristic of the samples in sample space RS based on DGA,a new self-adapted weight fuzzy omean clustering model of fault diagnosis of the power transformer based on the potential function is proposed. Meanwhile,from the aspect of geometry characteristic of FC-divided in s dimension sample space,a method is proposed for the purpose of getting an effective adjacent radius,adaptive cluster number c and original cluster center of X sample set. For the diagnosis
sample x,the property measure and diagnosis rule are proposed,which under the condition of potential density function that determine c number of optimal fuzzy cluster P1.
Plenty of diagnosis examples demonstrate the validity of the gray relationship analysis method in diagnosing power system insulation faults. Not only the misjudgment rate can be reduced,but also the transformer fault incidences can be classified by using self-adapted weight fuzzy clustering method. Further more,the transformer insulation condition can be analyzed,as well as the fault position can be located by recognizing the fault types. The diagnosing precision is much higher comparing with former three-ratio method.
引文
[1] 电科院．1995年度全国变压器、互感器、电抗器运行统计(110kV及以上电压等级)．过电压与绝缘．1997(3)：1～4
[2] 电科院．1996年全国变压器运行统计．过电压与绝缘．1998(7)：9
[3] 电气学会．变压器预防保全技术现状动向．电气学会技术报告(IIP)第334号．1990(8)：3～26
[4] 操敦奎．变压器油中气体分析与诊断．中国水利电力企业管理协会、武汉电力企业管理协会．1987
[5] Church J O et al. Analyze Incipient Faults with Dissolved-Gas Nomograph. Electrical world. 1987(10):40～44
[6] Kawamura T, et al. Dissolved Gas Analysis, its use for the Maintenance of Transformers. CIGRE paper. 12-05. 1986
[7] 杜洋．用“无编码比值法”分析和判断变压器故障性质．变压器．1999．36(3)：32～36
[8] 高文胜，钱政，杨莉，严璋．充油电力变压器氢气主导型故障的相关分析法．电网技术．1998．22(12)：55～58
[9] 李义仓．用色谱法诊断变压器过热故障及其部位．变压器．1995(7)：35～38
[10] 陈贻源．模糊数学．武汉：华中工学院出版社．1984
[11] 孙才新，郭俊峰，郑海平等．基于行为的变压器油色谱分析模糊诊断专家系统研究．电工技术学报．2001．16(3)：49～52，66
[12] 张鸣柳，孙才新．变压器油中气体色谱分析中以模糊综合评判进行故障诊断的研究．电工技术学报．1998．1(3)：51～54
[13] 孙才新，郭俊峰，廖瑞金，杜林，陈伟根．变压器油中溶解气体分析中的模糊模式多层聚类故障诊断方法的研究．中国电机工程学报．2001．21(2)：37～41
[14] 王平．基于模糊综合评判的变压器故障诊断法．电力系统自动化．1996．20(12)：30～33
[15] 何跃英，江荣汉．基于模糊理论的电力变压器故障诊断专家系统．电工技术学报．1994．9(3)：43～46
[16] Yann-Chang Huang, Hong-Tzer Yang, et al.Developing a New Transformer Fault Diagnosis System through Evolutionary Fuzzy Logic, IEEE Trans on Power Delivery, 1997,12(2):761～767
[17] 徐文，王大忠，周泽存．电气设备故障诊断中模糊性处理方法的探讨．高电压技术．1995．21(3)：46～48
[18] 钱政，杨莉，张冠军，严璋．基于模糊推理与模糊集理论的电力变压器故障诊断方法．电工电能新技术．1999(3)：36～39
[19] 姚陈果，廖瑞金，陈明英，许道林．故障诊断的面向对象知识库和专家系统．高电压技术．2001．27(1)：22～24
[20] 刘有才，刘增良．模糊专家系统原理与设计．北京：北京航空航天大学出版社．1995
[21] 宋斌，于萍，罗运柏，彭春华．变压器油状态监测的专家系统及应用研究．变压器．1998．35(12)：27～30
[22] 杨启平，薛玉德，符杨，蓝之达．专家系统在变压器故障诊断中的应用．变压器1996(6)：35～38
[23] 禹成七，张永浩，丰田淳一．基于ANN状态识别系统的变压器故障检测方法．华北电力大学学报．
1998．25(3)：19～23
[24] 王财政，孙才新．变压器色谱监测中的BPNN故障诊断法．中国电机学报．1997．17(5)：322
[25] 耿立思，潘旭峰，李晓雷，祝嘉光．灰色系统理论在机械故障诊断决策中的应用．北京理工大学学报．1997．17(2)：194～197
[26] Luo, M. F. Fault Detection, Diagnosis and Prognosis Using Grey System Theory. Monash University, A Dissertation Submitted for the Degree of Ph.D in Engineering,1994
[27] 施国洪，姚冠新．灰色系统理论在故障诊断决策中的应用．系统工程理论与实践．2001(4)：120～123，144
[28] 赵晖．用灰色系统法预测华东电网2005年最大峰谷差．华东电力．1999(9)：7～8
[29] X. Dong, Z. Wang, A. Kahtib, B. Qiu, Y. Liu Internet Based Fault Diagnosis for Power Transformers. IEEE PES Summer Meeting proceedings, Seattle, July 2000, 6 pages
[30] X. Dong, Y. Liu, et al. Internet Applications in Condition Monitoring of Power Equipment. the first International Conference on Insulation Condition Monitoring of Electrical Plants, Wuhan, China, June. 2000,10 pages
[31] X. Ding, E. Yao, Y. Liu, P. Griffin. ANN Based Analysis of Transformer Failures Using Gas-in-oil Data. Proceedings of The 55th American Power Conference, Chicago, IL ,April 1993,5 pages
[32] X. Ding, Y. Liu, An Artifical Neural Network Approach to Transformer Fault Diagnosis. Automation of Electric Power Systems (magazine), Jan. 1996, pp. 27～32,6 pages
[33] Zhenyuan Wang, Nien-chung Wang, Tzong-Yih Guo, Frank T C. Huang, Yilu-Liu, Paul J. Griffin. Artificial Intelligence in Power Equipment Fault Diagnosis. IEEE PES Winter Meeting Proceedings, Singapore, Feb. 2000,6 pages
[34] Zhenyuan Wang, Yilu Liu, Paul J. Griffin. Artificial Intelligence in OLTC Fault Diagnosis Using Dissolved GAS-In-Oil Information. IEEE PES Summer Meeting Proceedings,Seattle,WA,July 2000,6 pages
[35] 王道平，张义忠著．故障智能诊断系统的理论与方法．北京：冶金工业出版社．2001
[36] X. Dong, Z. Wang, Y. Liu, D. Zhu, K. Tan, F. Li, D. Lin, N. Wang, T. Guo. Internet Based Virtual Hospital Architecture For Power Equipment. IEEE PES Winter Meeting proceedings, Columbus, OH, Feb. 2000, 6 pages
[37] X. Dong, Z. Wang, B. Qiu, A. Khatib, Y. Liu, D. Zhu, K. Tan, F. Li, D. Lin, N. Wang, T. Guo. Virtual Hospital for Power Equipment on Iaternet. IEE APSCOM 2000, Hong Kong, March. 2000, 5 pages
[38] X. Dong, Z .Wang, Y. Lin, N. Wang, T. Guo, P. Grifin. Equipment Diagnosis and virtual Hospital Over the Internet. Proceedings of EPRI Diagnosis Couference, Feb. 2000, New Orleans. 11 pages
[39] 王昌长，李福祺．电力设备的在线监测与故障诊断．北京：清华大学出版社．1996
[40] 国家标准局．GB7252-87变压器油中溶解气体分析和判断导则．北京：中国标准出版社．1987
[41] 路长柏 编著．电力变压器绝缘技术．哈尔滨：哈尔滨工业大学出版社．1998
[42] 赵家礼，张庆达．变压器故障诊断与修理．北京：机械工业出版社．1998
[43] 董其国 编著．电力变压器故障与诊断．北京：中国电力出版社．2001
[44] 严璋．电气绝缘在线检测技术．北京：中国电力出版社．1995
[45] 张涵垺，何正嘉．模糊诊断原理及应用．西安：西安交通大学出版社．1992
[46] 吴今培 编著．模糊诊断理论及其应用．北京：科学出版社．1995
[47] 查键禄，田钦漠，李盛德．综合评价问题的系统分析．系统工程学报．2000．15(2)：124～130
[48] 冯圣洪，周心权．一种综合评价指标属性值规—量化技术新途径．系统工程学报．2000．15(2)：187～191
[49] 张冠军，严璋，张化君．电力变压器故障诊断中新方法的应用．高压电器．1998(4)：32～35
[50] Dunn J C. A Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact, Well Separate Clusters J. Cybern, 1974(3):32～57
[51] Bezdek J C. Pattern Recognition with Fuzzy Objective Function Algorithms, Plenum. New York, 1981
[52] 张冠军，钱政，严璋．变压器绝缘诊断中的模糊ISODATA法．高电压技术．1999．25(1)：1～3
[53] Pei Jihong, Fan Juitun. xie Weixin. Potential function Partial Weighted Fuzzy C-Mean(PWFCM) Clustering Method. International Conference on Signal Processing, ICSP'98,Beijing,china, 1998(10): 1209～1212
[54] 刘宜平，沈毅，刘志言．一种FCM聚类算法的改进与优化．系统工程与电子技术．2000．22(4)：1～3
[55] 高新波，裴继红，谢维信．模糊c—均值聚类算法中加权指数m的研究．电子学报．2000．28(4)：80～83
[56] N. R. Pal and J. C. Bezdek. On cluster validity for the fuzzy c—means model. IEEE Trans. Fuzzy Systems. 1995.3(3):370～379
[57] Ronald R. Yager and Dimitar P Filev. Approximate Clustering Via the Mountain Method. IEEE Trans. on SMC. Vol24. NO.8. August, 1994
[58] 孙才新等．变压器油中溶解气体的在线监测研究．电工技术学报．1996(2)：12～15
[59] 廖瑞金，陈伟根等．变压器油中溶解气体色谱的微机在线监测．高电压技术．1996(3)：30～32
[60] 唐炬，孙才新，李剑．电气设备绝缘在线监测传感器的研究．高压电器．1998(1)：13～18
[61] 陈伟根，孙才新，廖瑞金．变压器局部放电与油中气体浓度的关系．重庆大学学报(自然科学版)．2000．23(4)：86～89
[2] 电科院．1996年全国变压器运行统计．过电压与绝缘．1998(7)：9
[3] 电气学会．变压器预防保全技术现状动向．电气学会技术报告(IIP)第334号．1990(8)：3～26
[4] 操敦奎．变压器油中气体分析与诊断．中国水利电力企业管理协会、武汉电力企业管理协会．1987
[5] Church J O et al. Analyze Incipient Faults with Dissolved-Gas Nomograph. Electrical world. 1987(10):40～44
[6] Kawamura T, et al. Dissolved Gas Analysis, its use for the Maintenance of Transformers. CIGRE paper. 12-05. 1986
[7] 杜洋．用“无编码比值法”分析和判断变压器故障性质．变压器．1999．36(3)：32～36
[8] 高文胜，钱政，杨莉，严璋．充油电力变压器氢气主导型故障的相关分析法．电网技术．1998．22(12)：55～58
[9] 李义仓．用色谱法诊断变压器过热故障及其部位．变压器．1995(7)：35～38
[10] 陈贻源．模糊数学．武汉：华中工学院出版社．1984
[11] 孙才新，郭俊峰，郑海平等．基于行为的变压器油色谱分析模糊诊断专家系统研究．电工技术学报．2001．16(3)：49～52，66
[12] 张鸣柳，孙才新．变压器油中气体色谱分析中以模糊综合评判进行故障诊断的研究．电工技术学报．1998．1(3)：51～54
[13] 孙才新，郭俊峰，廖瑞金，杜林，陈伟根．变压器油中溶解气体分析中的模糊模式多层聚类故障诊断方法的研究．中国电机工程学报．2001．21(2)：37～41
[14] 王平．基于模糊综合评判的变压器故障诊断法．电力系统自动化．1996．20(12)：30～33
[15] 何跃英，江荣汉．基于模糊理论的电力变压器故障诊断专家系统．电工技术学报．1994．9(3)：43～46
[16] Yann-Chang Huang, Hong-Tzer Yang, et al.Developing a New Transformer Fault Diagnosis System through Evolutionary Fuzzy Logic, IEEE Trans on Power Delivery, 1997,12(2):761～767
[17] 徐文，王大忠，周泽存．电气设备故障诊断中模糊性处理方法的探讨．高电压技术．1995．21(3)：46～48
[18] 钱政，杨莉，张冠军，严璋．基于模糊推理与模糊集理论的电力变压器故障诊断方法．电工电能新技术．1999(3)：36～39
[19] 姚陈果，廖瑞金，陈明英，许道林．故障诊断的面向对象知识库和专家系统．高电压技术．2001．27(1)：22～24
[20] 刘有才，刘增良．模糊专家系统原理与设计．北京：北京航空航天大学出版社．1995
[21] 宋斌，于萍，罗运柏，彭春华．变压器油状态监测的专家系统及应用研究．变压器．1998．35(12)：27～30
[22] 杨启平，薛玉德，符杨，蓝之达．专家系统在变压器故障诊断中的应用．变压器1996(6)：35～38
[23] 禹成七，张永浩，丰田淳一．基于ANN状态识别系统的变压器故障检测方法．华北电力大学学报．
1998．25(3)：19～23
[24] 王财政，孙才新．变压器色谱监测中的BPNN故障诊断法．中国电机学报．1997．17(5)：322
[25] 耿立思，潘旭峰，李晓雷，祝嘉光．灰色系统理论在机械故障诊断决策中的应用．北京理工大学学报．1997．17(2)：194～197
[26] Luo, M. F. Fault Detection, Diagnosis and Prognosis Using Grey System Theory. Monash University, A Dissertation Submitted for the Degree of Ph.D in Engineering,1994
[27] 施国洪，姚冠新．灰色系统理论在故障诊断决策中的应用．系统工程理论与实践．2001(4)：120～123，144
[28] 赵晖．用灰色系统法预测华东电网2005年最大峰谷差．华东电力．1999(9)：7～8
[29] X. Dong, Z. Wang, A. Kahtib, B. Qiu, Y. Liu Internet Based Fault Diagnosis for Power Transformers. IEEE PES Summer Meeting proceedings, Seattle, July 2000, 6 pages
[30] X. Dong, Y. Liu, et al. Internet Applications in Condition Monitoring of Power Equipment. the first International Conference on Insulation Condition Monitoring of Electrical Plants, Wuhan, China, June. 2000,10 pages
[31] X. Ding, E. Yao, Y. Liu, P. Griffin. ANN Based Analysis of Transformer Failures Using Gas-in-oil Data. Proceedings of The 55th American Power Conference, Chicago, IL ,April 1993,5 pages
[32] X. Ding, Y. Liu, An Artifical Neural Network Approach to Transformer Fault Diagnosis. Automation of Electric Power Systems (magazine), Jan. 1996, pp. 27～32,6 pages
[33] Zhenyuan Wang, Nien-chung Wang, Tzong-Yih Guo, Frank T C. Huang, Yilu-Liu, Paul J. Griffin. Artificial Intelligence in Power Equipment Fault Diagnosis. IEEE PES Winter Meeting Proceedings, Singapore, Feb. 2000,6 pages
[34] Zhenyuan Wang, Yilu Liu, Paul J. Griffin. Artificial Intelligence in OLTC Fault Diagnosis Using Dissolved GAS-In-Oil Information. IEEE PES Summer Meeting Proceedings,Seattle,WA,July 2000,6 pages
[35] 王道平，张义忠著．故障智能诊断系统的理论与方法．北京：冶金工业出版社．2001
[36] X. Dong, Z. Wang, Y. Liu, D. Zhu, K. Tan, F. Li, D. Lin, N. Wang, T. Guo. Internet Based Virtual Hospital Architecture For Power Equipment. IEEE PES Winter Meeting proceedings, Columbus, OH, Feb. 2000, 6 pages
[37] X. Dong, Z. Wang, B. Qiu, A. Khatib, Y. Liu, D. Zhu, K. Tan, F. Li, D. Lin, N. Wang, T. Guo. Virtual Hospital for Power Equipment on Iaternet. IEE APSCOM 2000, Hong Kong, March. 2000, 5 pages
[38] X. Dong, Z .Wang, Y. Lin, N. Wang, T. Guo, P. Grifin. Equipment Diagnosis and virtual Hospital Over the Internet. Proceedings of EPRI Diagnosis Couference, Feb. 2000, New Orleans. 11 pages
[39] 王昌长，李福祺．电力设备的在线监测与故障诊断．北京：清华大学出版社．1996
[40] 国家标准局．GB7252-87变压器油中溶解气体分析和判断导则．北京：中国标准出版社．1987
[41] 路长柏 编著．电力变压器绝缘技术．哈尔滨：哈尔滨工业大学出版社．1998
[42] 赵家礼，张庆达．变压器故障诊断与修理．北京：机械工业出版社．1998
[43] 董其国 编著．电力变压器故障与诊断．北京：中国电力出版社．2001
[44] 严璋．电气绝缘在线检测技术．北京：中国电力出版社．1995
[45] 张涵垺，何正嘉．模糊诊断原理及应用．西安：西安交通大学出版社．1992
[46] 吴今培 编著．模糊诊断理论及其应用．北京：科学出版社．1995
[47] 查键禄，田钦漠，李盛德．综合评价问题的系统分析．系统工程学报．2000．15(2)：124～130
[48] 冯圣洪，周心权．一种综合评价指标属性值规—量化技术新途径．系统工程学报．2000．15(2)：187～191
[49] 张冠军，严璋，张化君．电力变压器故障诊断中新方法的应用．高压电器．1998(4)：32～35
[50] Dunn J C. A Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact, Well Separate Clusters J. Cybern, 1974(3):32～57
[51] Bezdek J C. Pattern Recognition with Fuzzy Objective Function Algorithms, Plenum. New York, 1981
[52] 张冠军，钱政，严璋．变压器绝缘诊断中的模糊ISODATA法．高电压技术．1999．25(1)：1～3
[53] Pei Jihong, Fan Juitun. xie Weixin. Potential function Partial Weighted Fuzzy C-Mean(PWFCM) Clustering Method. International Conference on Signal Processing, ICSP'98,Beijing,china, 1998(10): 1209～1212
[54] 刘宜平，沈毅，刘志言．一种FCM聚类算法的改进与优化．系统工程与电子技术．2000．22(4)：1～3
[55] 高新波，裴继红，谢维信．模糊c—均值聚类算法中加权指数m的研究．电子学报．2000．28(4)：80～83
[56] N. R. Pal and J. C. Bezdek. On cluster validity for the fuzzy c—means model. IEEE Trans. Fuzzy Systems. 1995.3(3):370～379
[57] Ronald R. Yager and Dimitar P Filev. Approximate Clustering Via the Mountain Method. IEEE Trans. on SMC. Vol24. NO.8. August, 1994
[58] 孙才新等．变压器油中溶解气体的在线监测研究．电工技术学报．1996(2)：12～15
[59] 廖瑞金，陈伟根等．变压器油中溶解气体色谱的微机在线监测．高电压技术．1996(3)：30～32
[60] 唐炬，孙才新，李剑．电气设备绝缘在线监测传感器的研究．高压电器．1998(1)：13～18
[61] 陈伟根，孙才新，廖瑞金．变压器局部放电与油中气体浓度的关系．重庆大学学报(自然科学版)．2000．23(4)：86～89