Based On Embedded Gas Concentration Detection System Software Design And Concentration Detection Algorithm Study

In daily life, people live in houses or cars, air pollution is almost everywhere,human life and health is seriously threatened. People urgently need a detect system todetect and assess environmental air quality fast and accurate. On the other hand, in themedical field, the exhaled gas detection can be achieved non-invasive rapid earlyscreening of diseases, but to achieve this goal, we also need a fast, accurate gasdetection system with high sensitivity and strong selectivity. At present, the research anddevelopment of gas detection system become more and more the concern of scholars,and porphyrin sensor array (PSA) chip gas detection technology is a hot spot in recentyears in this research direction. Compared with traditional methods, this method hasmany obvious advantages, it can be able to meet the demand of the sensitive、fast andaccurate detection.In this paper, with the PSA chips as gas sensors, we developed a set of the PSAimage acquisition and processing software on the Linux platform, and successfullyported to ARM platform to build a complete gas detection system. Then, with ammoniato be detected, data acquisition and processing was executed, and the quantitativeanalysis model of the gas concentration was studied. The validity and correctness of themodel was validated by experiments.The content of this study mainly includes the following aspects:①With Unified Modeling Language UML, system requirements、system modeland the system design is described in detail.②Set up the embedded software development environment. Complete of thedevelopment of man-machine interface, and migrate it to the target device.③With the ammonia experiments as an example, with the developed gas detectionsystem as detecting instrument, we achieved PSA chip reaction image acquisition andprocessing. Detailed analysis of the various factors that affect the data collected.④Based on the maximum curvature point of the time-response curve, analysis andfeature extraction for the data collected was conducted, and established quantitativeconcentration detection model for ammonia, executed quantitative detection of gasconcentrations with this model. For the time response curves of delayed or advancedinterference on the measurement result, we select the panning method to reduce theinterference. At last the ammonia and acetone experiments was conducted to prove the validity of the model. Meanwhile we gave the scope of the model.