| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.25 MB | Adobe PDF |
Resumo(s)
In the domain of IoT(Internet of Things), there is a correlation between the efficiency
and effectiveness of a platform, and the quality of the data available for processing and
decision making. When a scenario of data collection from different types of sensors,
integrated into several devices that are distributed alongside the building, is presented.
And, considering that data is essential to monitor and control the level of actuation in the
building, the process becomes a matter of extreme importance.
To achieve this, ANNODE’s methodology is adopted for quality verification of each
measurement that arrives to the framework. The construction and utilization of neural net works are crucial points of this project, as well as, the correct application of dependability
concepts, and notions of efficiency in processes of outlier detection and fault mitigation.
Descrição
Tese de mestrado, Informática, Universidade de Lisboa, Faculdade de Ciências, 2022
Palavras-chave
Internet das Coisas qualidade dos dados deteção de falhas mitigação de falhas Teses de mestrado - 2022
