沙万里,陈军豪,赵春晖
SHA Wanli,CHEN Junhao,ZHAO Chunhui

• 1:浙江浙能嘉华发电有限公司
• 2:浙江大学控制科学与工程学院

摘要(Abstract)：

在发电厂热力系统实际运行过程中,发电机负荷的频繁变化会使得热力系统关键设备部分运行参数随时间发生改变,存在明显的非平稳特性,由此导致设备的变工况问题非常普遍,故障特征很容易被非平稳趋势所掩盖,给及时检测设备异常带来较大困难。针对该问题,对步进有序时段划分算法进行推广,提出了一种基于负荷轴的工况划分算法,准确自动划分火电关键设备负荷信号,从而识别出不同的工况。基于划分出的不同工况,分别建立主元分析监测模型,在应用时根据新来样本的负荷调用相应的监测模型,从而实现对非平稳过程的状态监测。所提出的划分方法以监测模型的性能优劣来指示工况划分,相比一般基于聚类算法的工况划分方法更具有合理性。该算法的有效性已在某发电厂一次风机的故障案例得到验证。
During the operation of the thermodynamic system of a power plant, the frequent load change causes variation of operating parameters of key equipment and obvious nonstationary operating characteristics.As a result, the variable working conditions of the equipment are very common, and the fault characteristics are easily covered by nonstationary trends, which brings great difficulties to the timely detection of abnormalities of the equipment. To solve the problem, this paper generalizes the stepwise sequential phase partition(SSPP) algorithm and proposes a load axis based operating condition division algorithm to automatically divide the thermal key equipment load signals to identify different working conditions. Under the divided working conditions, the principal component analysis monitoring models are established respectively. In the application, the corresponding monitoring model is called according to the load of the new sample, thereby realizing the state monitoring of the nonstationary process. The division method proposed in this paper uses the performance of the monitoring model to indicate the division of working conditions, which is more reasonable than the general clustering method. Finally, the effectiveness of the proposed algorithm is verified in a failure case of a primary fan in a power plant.

关键词(KeyWords)： 非平稳过程;工况划分;状态监测;火力发电
nonstationary process;condition division;condition monitoring;thermal power generation

Abstract:

Keywords:

基金项目(Foundation): NSFC-浙江两化融合联合基金(U1709211);; 浙江省重点研发项目(2019C01048)

作者(Author): 沙万里,陈军豪,赵春晖
SHA Wanli,CHEN Junhao,ZHAO Chunhui

DOI: 10.19585/j.zjdl.201912006

参考文献(References)：

• [1]KOPPEN-SELIGER B,KIUPEL N,KELLINGHAUS H S,et al.A fault diagnosis concept for a high-pressure-pre heater line[C]//Proceedings of 1995 34th IEEE Conference on Decision and Control.1995(3):2383-2388.
• [2]BARSZCZ T,CZOP P.A feedwater heater model intended for model-based diagnostics of power plant installations[J].Applied Thermal Engineering,Elsevier Ltd,2011,31(8-9):1357-1367.
• [3]周东华,胡艳艳.动态系统的故障诊断技术[J].自动化学报,2009,35(6):748-758.
• [4]CHEN K Y,CHEN L S,CHEN M C,et al.Using SVM based method for equipment fault detection in a thermal power plant[J].Computers in Industry,Elsevier B.V.,2011,62(1):42-50.
• [5]WISE B M,RICKER N L,VELTKAMP D F,et al.Theoretical basis for the use of principal component models for monitoring multivariate processes[J].Process Control and Quality,1990,1(1):41-51.
• [6]黄孝彬,刘吉臻,牛玉广.主元分析方法在火电厂锅炉过程故障检测中的应用[J].动力工程学报,2004,24(4):542-547.
• [7]牛征,刘吉臻,牛玉广.基于主元分析的火电厂生产过程故障检测研究[J].华北电力大学学报(自然科学版),2005,32(4):33-37.
• [8]李晓杰,田亮.基于多尺度主元分析的磨煤机堵塞检测[J].电力科学与工程,2018,34(1):41-45.
• [9]张曦,阎威武,刘振亚,等.基于核主元分析和邻近支持向量机的汽轮机凝汽器过程监控和故障诊断[J].中国电机工程学报,2007,27(14):56-61.
• [10]ZHANG S,ZHAO C,ZHAO S.Two-level modeling strategy based incipient fault detection for nonstationary industrial processes[C]//2017 IEEE 56th Annual Conference on Decision and Control,CDC 2017.
• [11]CHEN Q,KRUGER U,LEUNG A Y T.Cointegration testing method for monitoring nonstationary processes[J].Industrial and Engineering Chemistry Research,2009,48(7):3533-3543.
• [12]LI G,QIN S J,YUAN T.Nonstationarity and cointegration tests for fault detection of dynamic processes[M].IFAC Proceedings Volumes(IFAC-PapersOnline),IFAC,2014.
• [13]ZHAO C,HUANG B.A full-condition monitoring method for nonstationary dynamic chemical processes with cointegration and slow feature analysis[J].AIChE Journal,2018,64(5):1662-1681.
• [14]许仙珍,谢磊,王树青.基于GMM的多工况过程监测方法[J].计算机与应用化学,2010,27(1):17-22.
• [15]李文卿.数据驱动的复杂工业过程统计过程监测[D].杭州：浙江大学,2018.
• [16]ZHAO C,SUN Y.Step-wise sequential phase partition(SSPP)algorithm based statistical modeling and online process monitoring[J].Chemometrics and Intelligent Laboratory Systems,Elsevier B.V.,2013(125):109-120.
• [17]YUE H H.Reconstruction-based fault identification using a combined index[J].Industrial&Engineering Chemistry Research,2001,40(20):4403-4414.