Study On Electric Fire Warning Algorithm Bbased On Fuzzy Neural Network

With the rapid development of China's economy,the level of electrification and automation of production and life has been continuously improved,and the frequency of electrical fires has become higher and higher,which may cause a large number of casualties and huge property losses.The electrical fire warning system can reduce the hazards of electrical fires,and with the development of technologies such as the Internet of Things,the performance of electrical fire warning systems in hardware such as fire signal detection and data transmission has been greatly improved.The current electrical fire warning system mostly uses the intuitive threshold method for the treatment of fire signals.The early warning accuracy for electrical fires is low,and the anti-interference ability is weak,which cannot meet the real-time effective warning requirements.Conducting research on electrical fire early warning algorithm based on fuzzy neural network to effectively improve the timeliness and accuracy of early warning of electrical fire early warning system,reduce false positives,false negatives and delays,and have important theoretical significance and practical application for effective prevention of electrical fires value.This paper introduces the direct method of fire signal processing,method of fire signal processing and the intelligent fire signal processing algorithm.It compares the fire characteristics,advantages and disadvantages of the above fire detection signal processing algorithms.Based on the comparison of the advantages and disadvantages of various fire detection signal processing algorithms,the fuzzy neural network algorithm is applied to the electrical fire early warning system.The algorithm combines the fuzzy logic algorithm with the artificial neural network algorithm to learn from each other to improve reliability and accuracy of fire warning systems.The MATLAB software is used to simulate the electrical fire early warning model based on neural network.The BP neural network-based electrical fire early warning algorithm model is trained by training data samples.The detected sample values are input into the trained network model,and the electrical fire probability of the actual output is compared with the expected electrical fire probability.When the probability of electrical fire output from the neural network early warning model is around 0.5,it is impossible to clearly determine whether an electrical fire will occur.In order to accurately predict the occurrence of electrical fires and improve the early warning accuracy of the system,a fuzzy is added after the neural network early warning model.Through the further logical judgment of the output result of the neural network early warning model,the fire situation judgment which is more in line with the actual situation is obtained,and the performance of the early warning model to accurately predict the electrical fire is verified by simulation experiments.