Indoor Fire Prediction Based On the CEEMDAN-LSTM-CNN Model

Gas indicators such as carbon monoxide, liquefied petroleum gas (LPG) and compressed natural gas (CNG) are often affected and changed during indoor fire events. However, the prediction accuracy is reduced due to the small trend of these metrics and the insignificant difference between the extreme values and minima. To solve this problem, a coupled model for fire indicators based on complete ensemble empirical modal decomposition (CEEMDAN), convolutional neural network (CNN), and long and short-term memory neural network (LSTM) is proposed. This model was applied to the prediction of gas indicators affected by fire, the prediction results of this model were compared with those of other models. The experimental findings demonstrate enhanced accuracy in predicting changes in gas indicators and a closer alignment with actual results through our model, offering a potent solution for indoor fire prevention.