A fuzzy neural network approach to fire detection in ships

This paper investigated a fuzzy neural network approach of fire detection in ships in order to detect a fire at the earlier stage and then give a high reliable judgement result. Conventional ship fire alarm systems often make simple logic judgement based on single sensor. To make a less erroneous alarm, two fire parameters (temperature and smoke density) are used in this approach. The processing of the sensor signals is carried out by fuzzy inference system based on neural networks. The simulation results demonstrate the effectiveness of this method.