摘要
为了解决压气机故障预警问题,通过对压气机典型运行特性进行分析,提出了一种基于多元状态估计技术(MSET)和偏离度的故障预警方法,并利用MSET建立了压气机正常运行状态下的非参数模型。对观测向量进行最优估计并得到估计向量,观测向量与估计向量之间的差异可以定性反映压气机是否异常。引入偏离度定量衡量观测向量与估计向量之间的偏离程度,然后利用滑动窗口法确定故障预警阈值,当平均偏离度超过预警阈值时,发出预警信息。实验结果表明,该方法有较高的预测精度,可以及时发现压气机的异常,从而实现对压气机运行状态的预警监测,减少经济损失。
In order to solve the problem of compressor failure warning,by analyzing the typical compressor operating characteristics,a fault warning method based on multivariate state estimation technology(MSET)and deviation degree is proposed,and a non-parametric model of the compressor under normal operating conditions is established by using MSET.The observation vector is optimally estimated and the estimated vector is obtained.The difference between the observation vector and the estimated vector can qualitatively reflect whether the compressor is abnormal.The degree of deviation is introduced to quantitatively measure the degree of deviation between the observed vector and the estimated vector,and then the sliding window method is used to determine the fault warning threshold.When the average degree of deviation exceeds the warning threshold,a warning message is issued.The experimental results show that the method has high prediction accuracy and can detect compressor abnormalities in time,realizing the early warning monitoring of the compressor operating status and reducing economic losses.
作者
陆永卿
涂雷
茅大钧
LU Yongqing;TU Lei;MAO Dajun(Shanghai Huadian Fengxian Thermal Power Co.,Ltd.,Shanghai 201403,China;Shanghai University of Electric Power,Shanghai 200090,China)
出处
《上海电力大学学报》
CAS
2021年第2期133-137,共5页
Journal of Shanghai University of Electric Power
基金
上海市“科技创新行动计划”地方院校能力建设专项项目(19020500700)
中国华电集团有限公司2020年度科技项目(CHDKJ20-02-149)。
关键词
压气机
多元状态估计
故障预警
滑动窗口法
compressor
multivariate state estimation technology
fault warning
sliding window method
