基于融合相似度的制粉系统健康预警及故障诊断

doi:10.16182/j.issn1004731x.joss.201802028

摘要/Abstract

摘要： 反映火电厂制粉系统健康状况的参数众多,同时监视这些参数难度较大且故障诊断过程过于复杂。为此提出了一种适合工业系统的基于融合相似度的健康预警实现方法和故障诊断算法。此融合相似度由一种新型的基于主成分分析的加权马氏距离和加权正弦相似度融合而成。基于该方法形成的诊断系统无需大规模样本库或记忆矩阵并具有自学习能力,算法运行中可自行修正计算融合相似度所需要的中心参数。仿真结果表明,该方法结构简单可靠,诊断正确率高,实时性强,适合在线运用。

关键词: 故障诊断, 相似度, 制粉系统, 健康预警, 马氏距离

Abstract: It is an arduous task to know health statusof a pulverizing system in power plant by monitoring these parameters simultaneously and the fault diagnosing process is complicated.An approach based on syncretic similarity is presented which is suitable for industrial system’s health warning and fault diagnosis.The syncretic similarity couples anew type of weighted mahalanobis distance based on principal component analysiswith an improved weighted sine similarity. The approachhas self-learning ability.Central parameterswhich are used to compute similarity can be modified along with the operation process.Simulation resultsshow that the method is suitable for online application because of its high accuracy, fast classification, high real-time performance, reliabilityand simple structure.

Key words: fault diagnosis, similarity, pulverizing system, health warning, mahalanobis distance

中图分类号:

焦嵩鸣. 基于融合相似度的制粉系统健康预警及故障诊断[J]. 系统仿真学报, 2018, 30(2): 595-604.

JiaoSongming. Health Warning and Fault Diagnosisof Pulverizing System Based on Syncretic Similarity[J]. Journal of System Simulation, 2018, 30(2): 595-604.

参考文献 17

[1]	陈蔚, 贾民平. 距离函数分类法在制粉系统故障诊断中的应用[J]. 振动、测试与诊断, 2009, 29(3): 282-286.CHEN Wei, JIA Minping. Application of Distance Function to Fault Diagnosis of Coal Pulverizing System [J]. Journal of Vibration, Measurement& Diagnosis, 2009, 29(3): 282-286.
[2]	刘鑫沛, 翟永杰, 张君颖. 基于聚类分析和状态估计的制粉系统故障预警[J]. 计算机仿真, 2013, 30(8):151-154.LIU Xinpei,ZHAI Yongjie,ZHANG Junying. Fault Warning of Pulverizing System Based onCluster Analysis and State Estimation[J]. Computer Simulation, 2013, 30(8):151-154
[3]	赵刚, 于向军, 吕震中. 运用RBF神经网络的制粉系统故障诊断[J]. 发电设备, 2006, 20(6): 449-453.ZHAO Gang, YU Xiangjun, Lv Zhenzhong. Diagnosis with RBF Neural Networks of Faults Occurring in Pulverizing Systems[J]. Power Equipment, 2006, 20(6): 449-453.
[4]	江若玫, 龚春琼. 基于极化因子神经网络的火电厂制粉系统故障诊断技术[J]. 计算机测量与控制, 2015, 23(5):1476-1478.Jiang Ruomei, Gong Chunqiong. Neural Network with Polarization Factor for Pulverizing System Fault Diagnosis[J]. Computer Measurement & Control. 2015, 23(5): 1476-1478.
[5]	费树岷, 李延红, 柴琳. 基于RSPNN的制粉系统故障诊断[J]. 控制工程, 2012, 19(3): 412-415.FEI Shumin,LI Yanhong,CHAI Lin. Fault Diagnosis Based on RSPNN for Pulverizing Systems[J]. Control Engineering of China. 2012, 19(3):412-415
[6]	Guo P, Infield D.Wind turbine tower vibration modeling and monitoring by the nonlinear State Estimation Technique (NSET)[J]. Energies (S1996-1073), 2012, 5(12): 5279-5293.
[7]	郭鹏, Infield David, 杨锡运. 风电机组齿轮箱温度趋势状态监测及分析方法[J]. 中国电机工程学报, 2011, 31(32): 129-136.GUO Peng, DAVID Infield, YANG Xiyun. Wind Turbine Gearbox Condition Monitoring Using Temperature Trend Analysis[J]. Proceedings of the CSEE, 2011, 31(32): 129-136.
[8]	肖先勇, 崔灿, 汪洋, 等. 电压暂降分类特征可比性,相关性及马氏距离分类法[J]. 中国电机工程学报, 2015, 35(6):1299-1305.XIAO Xianyong, CUI Can, WANG Yang, et al. Comparability and Relevance of Voltage Sag Classifying Features and Classification Methods Based on Mahalanobis Distance[J]. Proceedings of the CSEE, 2015, 35(6):1299-1305.
[9]	韩涵, 王厚军, 龙兵, 等. 基于改进马氏距离的模拟电路故障诊断方法[J]. 控制与决策, 2013, 28(11): 1713-1722.HAN Han, WANG Houjun, LONG Bing, et al. Method for Analog Circuit Fault Diagnosis Based on Improved MahalanobisDistance[J]. Control and Decision, 2013, 28(11): 1713-1722.
[10]	张胜君, 陈后金, 李艳凤, 等. 乳腺X线图像结构扭曲检测的研究[J]. 自动化学报, 2014, 40(8): 1764-1772.ZHANG Shengjun, CHEN Houjin, LI Yanfeng, et al. Detection of Architectural Distortion in Mammograms[J]. ACTA AUTOMATICA SINICA, 2014, 40(8): 1764-1772
[11]	张可, 周东华, 柴毅. 复合故障诊断技术综述[J]. 控制理论与应用, 2015, 32(9): 1143-1156. ZHANG Ke, ZHOU Donghua, CHAI Yi. Review of Multiple Fault Diagnosis Methods[J]. Control Theory & Applications, 2015, 32(9): 1143-1156
[12]	Z Gao, C Cecati, S X Ding.A survey of fault diagnosis and fault-tolerant techniques: Fault diagnosis with model- based and signal-based approaches[J]. IEEE Trans. Ind. Electro (S0278-0046), 2015, 62(6): 3757-3774.
[13]	陈义, 邹祥杰. 600 MW机组锅炉制粉系统故障原因分析及处理对策[J]. 华电技术, 2014, 36(7): 21-23.CHEN Yi, ZOU Xiangjie. Analysis on Fault Reasons of Coal Pulverizing System of 600 MW Boiler and Treatment CounterMeasures[J]. Huadian Technology, 2014, 36(7): 21-23.
[14]	樊帅, 肖军. 锅炉制粉系统故障诊断方法[J]. 热力发电, 2015, 44(2): 13-18. FAN Shuai, XIAO Jun. Fault Diagnosis Method of Pulverizing System in Utility Boiler[J]. Thermal Power Generation, 2015, 44(2):13-18
[15]	齐敏芳, 付忠广, 景源, 等. 基于信息熵与主成分分析的火电机组综合评价方法[J]. 中国电机工程学报, 2013, 33(2): 58-64.QI Minfang, FU Zhongguang, JING Yuan, et al. A Comprehensive Evaluation Method of Power Plant Units Based onInformation Entropy and Principal Component Analysis[J]. Proceedings of the CSEE, 2013, 33(2): 58-64.
[16]	沈媛媛,严严, 王菡子. 有监督的距离度量学习算法研究进展[J]. 自动化学报, 2014, 40(12): 2673-2686.SHEN Yuanyuan, YAN Yan, WANG Han-Zi.Recent Advances on Supervised Distance Metric Learning Algorithms[J]. ACTA AUTOMATICA SINICA, 2014, 40(12): 2673-2686.
[17]	Seungryong Kim, Bumsub Ham, Bongjoe Kim, et al.Mahalanobis distance cross correlation for illumination invariant stereo matching circuits and systems for video technology[J]. IEEE Trans. Circuits Syst. Video Technol (S1051-8215), 2014, 24(11): 1844-1859.