The Research of Gas Detection Based on data Fusion

Abstract

Aiming at the problem of gas pollution in the environment, the traditional testing equipments often appear the phenomenon of misstatement and false positives. A real-time online monitoring system is designed based on the detecting and control of harmful gases. The lower computer system make the STM32F103 microcontroller as the core, PC as the monitoring center, combining the technology of multi-sensor data fusion, using self-adapting weighting data fusion algorithm and decision level data fusion algorithm based on fuzzy set theory. Through the upper machine people can not only understand the quality of the indoor environment in time, and the system can inhibit the drift and noise of sensors to some extent, improve the measuring accuracy of the monitoring system for harmful gas in the environment, improve system reliability, reduce system fuzziness, enhance the reliability of system and assure the system having good robustness. Introduction Due to the restriction of economic development level, at present, the development of our country environment monitoring technology is not mature. Monitoring equipment and technology is relatively backward and the single, low degree of automation, most of gas monitor mainly is field sampling and laboratory analysis instrument as the main technical means, this method is not only low efficiency, high cost, but also costs a lot of manpower and material resources, financial resources, and analysis is slow, cannot satisfy the needs of real-time analysis of gas environmental quality [1]. Even though the market appears on a lot of portable gas detector at present, but the single test species, poor anti-interference ability and the poor precision, which can't satisfy to detect the gas concentration accurately and automatically. Therefore, developing the harmful gas detection system combining data fusion technology which is integrating data acquisition and environmental monitoring has important significance. Working Principle of System The system consists of two parts, the first part is upper computer data processing and display system composed of monitoring software of the computer, the second part is based on single chip microcomputer and sensors as the core, and intelligent detection and control system composed of other peripheral equipment. PC can monitor the concentration of harmful gases as well as the operation condition at any time, which can divide environmental quality into some kinds of grades respectively, each as excellent, good, bad and worse through the groups of data fusion. When the level of harmful gases is poor, PC warns MCU red lights flashing, when the level of harmful gases is worse, the system opens the sound and light alarm circuit, reminding people of harmful gas concentration is too high. International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2015) © 2015. The authors Published by Atlantis Press 2358 Three sensors of the same type is used in this system to analyze. First of all, the STM32 uploads the data of gas concentration to the PC through the RS-485 which are detected by the sensors, upload number for five times, after reaching five times, the upper computer uses adaptive weighted data fusion to process data. Then the upper computer uses decision level data fusion based on fuzzy set theory to grade the gas quality according the processed fusion results. Finally, the analysis results will be shown by monitoring the interface and make corresponding alarm response according to the results. The main program flow chart of the system is shown in figure 1. start system initialization detection of gas concentration upload to the upper computer If upload number up to 5 times adaptive weighted data fusion algorithm decision level data fusion algorithm of fuzzy set theory