国产欧美精品一区二区,中文字幕专区在线亚洲,国产精品美女网站在线观看,艾秋果冻传媒2021精品,在线免费一区二区,久久久久久青草大香综合精品,日韩美aaa特级毛片,欧美成人精品午夜免费影视

In order to realize intelligent home system early warning and timely and accurate alarm to protect the home environment and personal property safety, a smart home fire early warning system is developed. In the early warning system was established by using BP neural network prediction model of fire, the Home Furnishing temperature, smoke concentration, the concentration of carbon monoxide gas - fire information as input parameters, fire occurrence probability, the probability of occurrence of smoldering fire and no fire occurrence probability as output, combined with the use of the LM algorithm and genetic algorithm for optimization of fire prediction model greatly. Experiments show that the prediction accuracy of the warning system is high, and it can effectively improve the low intelligence level of the traditional fire prediction.