Towards a Fire Alarm Model Based on Variable Learning Rate Algorithm with Weight Factor

DOI

10.2991/mce-14.2014.22How to use a DOI?

Keywords

fire alarm; neural network; variable learning rate; weight factor

Abstract

Fire alarm with Neural Network (NN) can learn knowledge from multiple sensor data fusion. By adjusting network weight, more stable fire alarm can be achieved. Classic BP NN is likely to fall into local minimum. To address this problem, a Variable Learning Rate Algorithm with Weight Factor (VLRA-BP) was proposed and introduced into automatic intelligent decision in fire alarm. The model with VLRA-BP algorithm uses temperature and smoke sensors to perform intelligent information transformation, so as to achieve target of timely alarming and decrease system false alarm rate. Simulation experiment result shows that system accuracy and average error can effectively monitor simulated fire scene and forecast fire.

Copyright

© 2014, 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/).

Download article (PDF)

TY  - CONF
AU  - Wei Wang
AU  - Liping Yang
AU  - Lei Xu
PY  - 2014/03
DA  - 2014/03
TI  - Towards a Fire Alarm Model Based on Variable Learning Rate Algorithm with Weight Factor
BT  - Proceedings of the 2014 International Conference on Mechatronics, Control and Electronic Engineering
PB  - Atlantis Press
SP  - 103
EP  - 106
SN  - 1951-6851
UR  - https://doi.org/10.2991/mce-14.2014.22
DO  - 10.2991/mce-14.2014.22
ID  - Wang2014/03
ER  -