Design Of Fire Protection System Based On Dual Environment Perception And Intelligent Processing

With the rapid development of society,the architectural style of buildings is constantly being updated,showing the phenomenon that the floors are getting higher and higher,the internal structure of the building is more and more complex,ndoor flammable and explosive materials and electrification equipment are obviously increased.Most of the existing fire monitoring and alarm systems have a single monitoring method,limited data sources,serious problems of underreporting and false positives.This has led to a sharp increase in fire hazards,and there is an urgent need for real-time monitoring and alarming of building fire protection environments.Aiming at the serious underreporting,false positives and the inability to adapt to the complex real environment of buildings,this paper designs a fire protection system based on dual environment sensing and intelligent processing.The main research contents include:(1)Research on multi-source heterogeneous fire data sensing method based on dual environment.Since the fire-fighting equipment plays an extremely important role in the fire,the system adds the fire-fighting gas cylinder to the fire-sensing environment,and combines the fire hazard points to form two fire data acquisition scenarios.Multisource heterogeneous data information includes cylinder pressure,ambient temperature and humidity,temperature,humidity,smoke,and image of fire hazard points.The multisource heterogeneous data of the fire protection environment has heterogeneous characteristics in terms of data types,data structures,and sensor communication protocols.This paper integrates the methods of sensing and acquiring various types of monitoring signals,and designs a set of software and hardware methods for multisource heterogeneous fire data monitoring.(2)Based on LoRa star polling nested networking method research.Faced with the complexity of the structure of today's buildings,the number of data collection nodes is increasing,the amount of data transmission is large,and the distance is long,the lowpower wide area network LoRa technology is used for real-time data transmission in multi-node networking.The data collision and packet loss problem of LoRa wireless transmission is solved by studying LoRa star network polling and data collision delay.This paper studies and proposes to use LoRa star network to poll the nested networking method to solve the upper limit of the number of child nodes in the star network polling networking mode,and improve the stability and robustness of LoRa wireless network transmission.(3)Fire data intelligent processing method based on multi-layer perceptron neural network.For the multi-source heterogeneous attributes of system environmentaware data,in order to mine data information and determine the state of the perceived environment in real time,this paper proposes to use multi-layer perceptron neural network algorithm to process fire data.By studying the algorithm learning of multilayer perceptron neural network,the algorithm model suitable for this system is established,and the algorithm model is trained and analyzed.In order to further improve the accuracy of the system alarm,the scene picture of the fire hazard point is flame-recognized,and the processing result of the multi-layer perceptron neural network is combined to accurately determine the fire occurrence and realize the intelligent processing of the fire data.Finally,according to the above research on the system scheme,select the relevant sensor,STM32 minimum system,LoRa wireless transceiver module,touch screen and 4G network module,design and manufacture the system hardware circuit and the receiving processing terminal display cabinet,and write the system algorithm program based on the hardware platform.The function of using the single-chip computer to control the touch screen to display the monitoring data information and accurately transmitting the alarm information to the smart phone terminal is realized.Through experimental tests,the dual environment awareness and intelligent processing of the fire protection system can meet the design requirements of real-time data collection,stable network transmission,accurate data processing and precise information delivery.