基于模糊神经网络火灾探测信号处理方法

Signal Processing Method for Fire Detection Based on Fuzzy Neural Network

针对火灾探测信号的特点,建立了火灾探测系统模型及用于处理火灾信号的模糊神经网络计算模型.利用神经网络构造模糊系统,用神经网络的自学习和自适应能力自动调整模糊系统参数,用改进的BP算法对网络进行学习和训练.根据国家标准试验火数据进行网络的学习和测试,系统误差小于试验火标准误差要求,表明了算法的有效性和可行性.

A fire detection system model and calculating model of fuzzy neural network for processing fire signal are proposed aiming at the characteristic of fire detection signal. Through integrating fuzzy system and neural network, this method realizes constructing fuzzy system by neural network and adjusting system parameter automatically by self-learning and self-adapting ability of neural network.. Using reformative BP algorithm completes the network learning and training. The network is trained and tested using country testing fire data, system error is less than standard error, which shows the validity and feasibility of the algorithm.