Publicação
Fault detection agents based on wavelet adaptive thresholds
| Resumo: | Designing a Fault Tolerant Networked Control System (FTNCS) to deal with large-scale complex networked control systems is a very difficult task due to the large number of sensors and actuators spatially distributed and networked connected. To deal with this problem, this paper presents a new agent based fault detection (FD) system for complex and dynamic processes, where the modularity and complexity of the processes are important aspects in the FD agent system constructed. The FD agents proposed here have hybrid architectures based in a horizontal layers, where the reactive layer is based on Wavelet Adaptive Thresholds. The results, using a distributed benchmark with a three tank process, show that the FD agents proposed are a good alternative to classic FD methods. |
|---|---|
| Autores principais: | Mendes, Mário J. G. C. |
| Outros Autores: | Costa, José Sá da |
| Assunto: | Fault tolerant networked control Fault detection Agents Wavelet Decomposition methods |
| Ano: | 2010 |
| País: | Portugal |
| Tipo de documento: | documento de conferência |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Universidade Politécnica de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Científico da Universidade Politécnica de Lisboa |
| Resumo: | Designing a Fault Tolerant Networked Control System (FTNCS) to deal with large-scale complex networked control systems is a very difficult task due to the large number of sensors and actuators spatially distributed and networked connected. To deal with this problem, this paper presents a new agent based fault detection (FD) system for complex and dynamic processes, where the modularity and complexity of the processes are important aspects in the FD agent system constructed. The FD agents proposed here have hybrid architectures based in a horizontal layers, where the reactive layer is based on Wavelet Adaptive Thresholds. The results, using a distributed benchmark with a three tank process, show that the FD agents proposed are a good alternative to classic FD methods. |
|---|