检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:郭金玉 王霞 李元 GUO Jinyu;WANG Xia;LI Yuan(College of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China)
机构地区:[1]沈阳化工大学信息工程学院,辽宁沈阳110142
出 处:《大连工业大学学报》2022年第6期452-461,共10页Journal of Dalian Polytechnic University
基 金:国家自然科学基金项目(62273242);辽宁省教育厅科学研究项目(LJ2019007).
摘 要:为了进一步提高动态监测模型的适用性和可靠性,需要对给定数据集的结构信息进行完整的保留,而不是单方面的保留。针对传统动态多元统计分析方法不能同时考虑给定数据集的不同结构特征问题,提出一种基于集成动态结构分析(EDSA)的故障检测算法。分别利用动态主元分析、动态局部保持投影和动态多流形投影模型计算其T 2和平方预测误差(SPE)统计量;通过贝叶斯推理将各个模型的统计量组合成一个集成统计量;通过集成统计量与控制限对比确定数据是否故障。通过数值例子和田纳西-伊斯曼过程来验证基于EDSA的故障检测算法的可行性和可靠性。仿真结果表明,EDSA的整体性能优于传统的单一算法。新算法解决了传统算法中无法保留数据集不同结构特征信息的问题。To further improve the applicability and reliability of the dynamic monitoring model,it is necessary to retain the structural information of a given dataset completely,rather than unilaterally.A fault detection algorithm based on ensemble dynamic structural analysis(EDSA)was proposed to solve the problem that the traditional dynamic multivariate statistical analysis method did not consider the different structural characteristics of a given data set at the same time.The dynamic principal component analysis,dynamic locality preserving projections and dynamic multi-manifold projections models were used to calculate T 2 and SPE statistics of each model.The statistics of each model were integrated into an ensemble statistics by Bayesian inference.It was determined whether the data was a fault by comparing the ensemble statistics with the control limits.The feasibility and reliability of the fault detection algorithm based on EDSA were verified through a numerical example and the Tennessee-Eastman process.Simulation results showed that the overall performance of EDSA was better than traditional single algorithm.The new algorithm solves the problem that the different structural information of data set cannot be retained in the traditional algorithm.
关 键 词:故障检测 动态主元分析 动态局部保持投影 动态多流形投影 集成学习 贝叶斯推理
分 类 号:TP277[自动化与计算机技术—检测技术与自动化装置]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.145