This paper addresses a problem of observer-based sensor fault reconstruction for continuous-time systems subject to sensor faults and measurement disturbances via a descriptor system approach. An augmented descriptor ...This paper addresses a problem of observer-based sensor fault reconstruction for continuous-time systems subject to sensor faults and measurement disturbances via a descriptor system approach. An augmented descriptor plant is first formulated, by assembling measurement disturbances and sensor faults into an auxiliary state vector. Then a novel descriptor state observer for the augmented plant is constructed such that simultaneous reconstruction of original system states, sensor faults and measurement disturbances are obtained readily. Sufficient conditions for the existence of the proposed observer are explicitly provided, and the application scope of the observer is further discussed. In addition, an extension of the proposed linear approach to a class of nonlinear systems with Lipschitz constraints is investigated. Finally, two numerical examples are simulated to illustrate the effectiveness of the proposed fault-reconstructing approaches.展开更多
Multimedia Services has drawn much attention from both industrial and academic researchers due to the emerging consumer market, how to provide High-Availability service is one of most important issues to take into acc...Multimedia Services has drawn much attention from both industrial and academic researchers due to the emerging consumer market, how to provide High-Availability service is one of most important issues to take into account. In this paper, a dynamic fault tolerant algorithm is presented for highly available distributed multimedia service, then by introducing SLB(server load balancing) into fault tolerance and switching servers in different ways according to their functions, the proposed schema can preserve reliability and real-time of the system .The analysis and experiments indicate that resuming server's faulty by this method is smooth and transparent to the client The proposed algorithm is effectively improving the reliability of the multimedia service.展开更多
Effective network management software ensures networks to run credibly. In this paper we discuss the design and implementation of network device fault management based on Pure Java. It includes designs of general func...Effective network management software ensures networks to run credibly. In this paper we discuss the design and implementation of network device fault management based on Pure Java. It includes designs of general functions, server functions, client functions and a database table. The software can make it convenient to monitoring a network device, and improve network efficiency.展开更多
This paper describes the development of the condition monitoring and faultdiagnosing system of a group of rotating machinery. The data management is performed by means ofdouble redundant data bases stored simultaneous...This paper describes the development of the condition monitoring and faultdiagnosing system of a group of rotating machinery. The data management is performed by means ofdouble redundant data bases stored simultaneously in both the analyzing server and monitoringclient. In this way, high reliability of the storage of data is guaranteed. Condensation of trenddata releases much space resource of the hard disk. Diagnosing strategies orientated to differenttypical faults of rotating machinery are developed and incorporated into the system. Experimentalverification shows that the system is suitable and effective for condition monitoring and faultdiagnosing for a rotating machine group.展开更多
针对服务器的故障检测精度较低导致服务器维护成本增加的问题,提出了一种利用大型语言模型实现服务器故障的检测方法。利用BERT(Bidirectional Encoder Representation of Transformer)大型语言模型对服务器运行状态文本进行语义分析,...针对服务器的故障检测精度较低导致服务器维护成本增加的问题,提出了一种利用大型语言模型实现服务器故障的检测方法。利用BERT(Bidirectional Encoder Representation of Transformer)大型语言模型对服务器运行状态文本进行语义分析,生成高维词向量,以充分捕捉文本中的语义信息。基于生成的词向量,计算各词向量的权重值和互信息值,筛选出对故障检测具有显著贡献的关键词向量,从而降低数据维度并提升特征提取的准确性。将筛选出的关键词向量作为输入,利用GG(Gaussian-Gamma)聚类算法进行聚类分析,通过迭代优化聚类中心和隶属度矩阵,将服务器运行状态划分为正常状态和故障状态,并进一步识别具体故障类型。实验结果表明,该方法在关键词向量提取和故障检测性能上均表现出色,能够有效提升服务器故障检测的精度和效率,为降低服务器维护成本提供了可靠的技术支持。展开更多
基金supported by the National Natural Science Foundation of China(61104026)the Open Funding for National Defense Key Subject Laboratory of Micro and Small Spacecraft Technology(20090450126)
文摘This paper addresses a problem of observer-based sensor fault reconstruction for continuous-time systems subject to sensor faults and measurement disturbances via a descriptor system approach. An augmented descriptor plant is first formulated, by assembling measurement disturbances and sensor faults into an auxiliary state vector. Then a novel descriptor state observer for the augmented plant is constructed such that simultaneous reconstruction of original system states, sensor faults and measurement disturbances are obtained readily. Sufficient conditions for the existence of the proposed observer are explicitly provided, and the application scope of the observer is further discussed. In addition, an extension of the proposed linear approach to a class of nonlinear systems with Lipschitz constraints is investigated. Finally, two numerical examples are simulated to illustrate the effectiveness of the proposed fault-reconstructing approaches.
文摘Multimedia Services has drawn much attention from both industrial and academic researchers due to the emerging consumer market, how to provide High-Availability service is one of most important issues to take into account. In this paper, a dynamic fault tolerant algorithm is presented for highly available distributed multimedia service, then by introducing SLB(server load balancing) into fault tolerance and switching servers in different ways according to their functions, the proposed schema can preserve reliability and real-time of the system .The analysis and experiments indicate that resuming server's faulty by this method is smooth and transparent to the client The proposed algorithm is effectively improving the reliability of the multimedia service.
文摘Effective network management software ensures networks to run credibly. In this paper we discuss the design and implementation of network device fault management based on Pure Java. It includes designs of general functions, server functions, client functions and a database table. The software can make it convenient to monitoring a network device, and improve network efficiency.
文摘This paper describes the development of the condition monitoring and faultdiagnosing system of a group of rotating machinery. The data management is performed by means ofdouble redundant data bases stored simultaneously in both the analyzing server and monitoringclient. In this way, high reliability of the storage of data is guaranteed. Condensation of trenddata releases much space resource of the hard disk. Diagnosing strategies orientated to differenttypical faults of rotating machinery are developed and incorporated into the system. Experimentalverification shows that the system is suitable and effective for condition monitoring and faultdiagnosing for a rotating machine group.
文摘针对服务器的故障检测精度较低导致服务器维护成本增加的问题,提出了一种利用大型语言模型实现服务器故障的检测方法。利用BERT(Bidirectional Encoder Representation of Transformer)大型语言模型对服务器运行状态文本进行语义分析,生成高维词向量,以充分捕捉文本中的语义信息。基于生成的词向量,计算各词向量的权重值和互信息值,筛选出对故障检测具有显著贡献的关键词向量,从而降低数据维度并提升特征提取的准确性。将筛选出的关键词向量作为输入,利用GG(Gaussian-Gamma)聚类算法进行聚类分析,通过迭代优化聚类中心和隶属度矩阵,将服务器运行状态划分为正常状态和故障状态,并进一步识别具体故障类型。实验结果表明,该方法在关键词向量提取和故障检测性能上均表现出色,能够有效提升服务器故障检测的精度和效率,为降低服务器维护成本提供了可靠的技术支持。
