The Research On Data Processing And Calibration Of Odorous Detecting System

As a special atmospheric pollution,odor pollution has seriously affected the people's production and life,and even threatens people's lives and safety.Odorous detection is an important basis for environmental governance and protection,but there are many defects in the odor detection instrument at present,such as the low detecting precision,the high price,off-line detection which lead to the poor real-time property,etc.Detection method of malodorous and instrument development is of great significance.In this thesis,data processing and calibration techniques of malodorous detecting system are studied.Firstly,this paper introduces the principle of the odorous detecting and the structure and work process of detecting system.The odorous detecting system consists of the gas system,the injection system,the separation system,the detection system,the control system,and the data processing system.Afterwards,data processing and calibration methods of detection system,which is the focus of this article,is introduced.According to the type of electrical noise superimposed in the odorous detecting system,the noise removal method is researched in order to improve the signal-to-noise ratio of detecting system.By comparing several filtering methods,wavelet threshold processing is selected to complete the noise elimination of GC data.According to the changing trend of the peak characteristic points in the curve,find the vertex position by using the pseudo derivative method firstly,and then look down on both sides from the apex of formula beginning and the end to find the characteristic points.For the calculation of chromatographic peak area,firstly determine the type of peak,and then the corresponding area calculation method is adopted depending on the type of peak shape and adjust the baseline to improve calculation accuracy.For the qualitative and quantitative analysis of materials,we use the query time window method to complete the identification of the various components in the gas mixture;by comparing several quantitative methods,external standard method and PID correction coefficient is adopted to complete the quantitative calculation of the malodorous,and the feasibility of the algorithm is verified by experiment.In this paper,the algorithm is verified and analyzed from two aspects of the simulation experiment and the actual experiment.Comsol is used to simulate the flow signal in the GC column and Matlab is used to realize the simulation analysis and comparison of the denoising algorithm in the simulation experiment stage.Afterwards,the reliability of the whole system is tested by the experiment of stability and repeatability on the system.The error of fluctuation caused by temperature,flow rate and other environmental changes is analyzed and compensated.Then,the separation efficiency and detection accuracy of gas components are verified by experiments.It is proved that the measurement precision is better than 6% and the minimum detection limit reaches 1ppm with good repeatability.Finally,the shortages and corresponding improvements of the system are also discussed in this paper.