A Remote Data Acquisition And Automatic Fire Alarm System

The rapid development of modern urban construction led to the thriving growth of the real estate industry, and high-level buildings have been continuously springing up in various size cities. High-level buildings not only bring convenience to our lives, but also bring a lot of fire hazards with them. In view of the shortcomings of the traditional monitoring system, such as difficult operation, high operating cost and difficult wiring,which can not satisfy the requirement of high-level buildings' fire safety. On the basis of the working principles and application characteristics of the traditional automatic fire alarm system as well as current level of communication technology. This article has developed a set of wireless, intelligent, strongly expansible, convenient query and maintenance monitoring system, which is based on ZigBee and GPRS and look forward to satisfy the requirement of high-level buildings.This design mainly includes three parts: data collection module, control center module and data monitoring module. The data collection module firstly collect the input signals from the indoor temperature, smoke and CO concentration sensors, which are then sent fire signals out via ZigBee module CC2530; the control center STM32F103VCT6 receives the radio frequency signal, and collects the real-time temperature, smoke particle concentration and CO concentration, which are all to be displayed through a LCD screen;Ultimately it comes down to data monitoring module. GPRS communication module and TCP/IP connection are designed and completed. Users can directly view the data collected and sent by the different detectors through the logic system to facilitate timely alarm. In this way, the system realizes the remote wireless data exchange between ZigBee network and the Internet, and the monitors and users can easily monitor remote fires without even stepping outside.The most important Innovation point of this article is the application of multi-sensor data fusion technology, which optimizes data level, feature level and decision level. Data level has adopt normalization method,amplitude limit and variational threshold algorithm;Feature level fusion has been based on the data from data level which actually adopt the method of BP neural network; On basis of the feature extract by feature level, decision level has used fuzzy control technology to get a reasonable and accurate decision output.The new system will make a great improvement in comparison with traditional simple threshold algorithm on veracity to reduce false positive rate and false negative rate.As field operation results would show, the automatic fire alarm system designed by this article can ensure long-term and steady operation,and it also has a friendly man-machine exchange interface, which basically satisfies the need to monitor modern buildings for fire safety. Compared with the traditional alarm system, it not only reduces the labour cost but realized wireless communications; it reduces the wiring cost and meanwhile it is of strong expansibility, convenient maintenance, simple operation.