원문정보
초록
영어
In accordance with the requirements for home security and safe guard, a new type of intelligent monitoring system is researched and developed. The system is established with CAN bus and wireless as the foundation. Multi-sensor technology is used to improve the alarm algorithm of the system, a new fuzzy neural network is put forward as the classifier. There are four layers in the network, and the input signals are temperature and the concentration of CO, smoke and CO2, the output signal is the fire probability. Radial basis function (RBF) is used as the fuzzy membership function. The principal component analysis (PCA) is used to extract the information of sensors and an observation window is used to extract the necessary information for neural network. Experiments show that the system can provide accurate detection results in a short time.
목차
1. Introduction
2. The Overall Hardware Structure of Monitoring System
3. Fire Alarm System Based on Data Fusion
3.1 Sensor Selection of Fire alarm System
3.2 PCA algorithm used for Feature Extraction
3.3 Extract the Characteristics by Fuzzy Neural Network
4. Data Selection and Experiment Analysis
5. Experimental Results and Analysis
5.1 Sensitivity Analysis
5.2 Reliability Analysis
6. Conclusion
References