Research on Fire Detection method of Substation Based on Multi-sensors information Dynamic Tracing and Fusion Technology
- DOI
- 10.2991/icsmim-15.2016.164How to use a DOI?
- Keywords
- Substation, Multi-sensors, Fire detection, Information Dynamic Tracing, Information Fusion, BP neutral network
- Abstract
The complex detectors with multi-sensors of feature parameters are developed for realizing early fire-detection in substation. The fire detecting signal measurements from these multi-sensors and their variation rates are used as the data of substation fire judgment. The time of fire alarm is advanced through tracing and calculating the variation rates of the detecting signals. The multi-detecting signals are processed and fused through Back Propagation (BP) neutral network, which not only avoids shortcomings of traditional fire detectors using a single fire signal as fire judgment but also reduces false positives and false negatives of fire. The training samples including training set and test set are built, and feedback correction for the samples ensures their rationality. In order to meet the stability of fire detection results in substation, batch mode for training is applied. Server push technology presented here accelerates speed of fire response further. Tests prove the feasibility and validity of the above mentioned ways, which means the accuracy and the speed of fire alarm in substation are greatly enhanced.
- Copyright
- © 2016, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Shuping Wang AU - Haichao Wang AU - Minghao Fan AU - Haicheng Wu AU - Jiaqing Zhang PY - 2016/01 DA - 2016/01 TI - Research on Fire Detection method of Substation Based on Multi-sensors information Dynamic Tracing and Fusion Technology BT - Proceedings of the 2015 4th International Conference on Sensors, Measurement and Intelligent Materials PB - Atlantis Press SP - 887 EP - 896 SN - 2352-538X UR - https://doi.org/10.2991/icsmim-15.2016.164 DO - 10.2991/icsmim-15.2016.164 ID - Wang2016/01 ER -