Microwave‐based system for non‐destructive monitoring water pipe networks using support vector machine
| Autor: | Henrique E. M. Peres, Fatima Salete Correra, Wesley Becari, Arthur M. de Oliveira |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Engineering
business.industry Acoustics 020208 electrical & electronic engineering 010401 analytical chemistry 02 engineering and technology 01 natural sciences Atomic and Molecular Physics and Optics 0104 chemical sciences Support vector machine Pipe network analysis Microstrip antenna Kernel (linear algebra) Position (vector) Q factor 0202 electrical engineering electronic engineering information engineering Electronic engineering Electrical and Electronic Engineering Antenna (radio) business Microwave |
| Zdroj: | IET Science, Measurement & Technology. 10:910-915 |
| ISSN: | 1751-8830 |
| DOI: | 10.1049/iet-smt.2016.0131 |
| Popis: | Water is the most valuable natural resource. To preserve this good, several water monitoring systems for pipe networks have been proposed. However, there is a lack of research on the development of high-frequency electromagnetic sensors for this task. This study proposes a non-destructive technique to locate carrying water polyvinyl chloride pipes using a microstrip patch antenna. An experimental apparatus was mounted and the resonant frequency and Q factor of the antenna were measured for different relative positions between the sensor and the pipe filled with water. To detect the pipe position, experimental data were trained and tested by applying an approach based on support vector machines. The best results were from using a simple linear kernel, which allowed 100% accuracy using a 30% fraction of data used for holdout validation. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text