Gas Detection System Based on Multi-Sensor Fusion with BP Neural Network

Using the sensor technology as the core, combining with the A/D technology, pattern recognition technology and other technologies, we have designed a new type of gas detection system. Gases are detected through BP neural network algorithm. The design of the system, the establishment of BP neural network and the improvement of its algorithm are given. The fusion technology of multi-sensor is expounded. Firstly, gases are detected by the gas sensor array. Secondly, the detected signals are sent to the preconditioning circuits of signals for corresponding process, and then they are sent to the pattern recognition circuit, signals are analyzed and recognized in the pattern recognition circuit. Finally, signals are input into the controlling circuits, relevant information are displayed or broadcast. Repeated experiments show that the gas detection system based on multi-sensor fusion with BP neural network can detect gases well, so it has higher application value. Copyright © 2013 IFSA.