Key Technology Research On Two-dimensional Distribution Of Ammonia Gas In Farmland Based On Tunable Laser

The excessive application of nitrogen fertilizers not only causes a reduction in the efficiency of nitrogen fertilizer use and the waste of resources,but also leads to various ecological and environmental problems.Ammonia volatilisation is an important pathway for nitrogen loss when nitrogen fertiliser has been applied to farmland and a major source of atmospheric ammonia,but excessive atmospheric ammonia deposition can lead to a reduction in photosynthetic rates,eutrophication in lakes or offshore and soil acidification.Therefore,quantitative analysis of changes in ammonia concentrations on farmland is significant for improving the efficiency of agricultural production and protecting the environment.The low concentration of ammonia in air,which is a trace gas,makes it difficult for current conventional analytical techniques to accurately achieve online monitoring of ammonia concentrations,limiting the level of research into ammonia detection in farmland.Tunable Diode Laser Absorption Spectroscopy(TDLAS)offers the advantages of high accuracy,rapid response and non-contact measurement.This thesis combines the TDLAS technique with a 2D image reconstruction algorithm to reconstruct the 2D distribution of ammonia in agricultural fields.The principal research elements and achievements are as follows:Based on the TDLAS technique,a new optical path structure for TDLAS detection is designed.Compared to conventional detection optical path structures,the detection optical path in this thesis offers both a 2D reconstruction of the ammonia concentration and the reduction of the number of optical paths,improving the applicability of the detection system.Based on the framework of the new optical path structure,an investigation is conducted into the impact of initial values and relaxation factors in image reconstruction algorithms on the resulting reconstructions.The simulation of the data shows that the setting of the initial values is closely related to the speed of image reconstruction and that the selection of the relaxation factor has a significant impact on the accuracy of the reconstructed image.This thesis proposes a noise reduction algorithm based on Variational Mode Decomposition(VMD)for TDLAS detection signals in large scale open space.The decomposition,filtering,filtering and reconstruction of the detection signal improve the signal-to-noise ratio of the detection signal,improve the fitting effect of the absorption signal without losing the effective signal,and improve the detection performance of the system.When the method was applied to the experimental signal,the fit of the obtained fitted curve improved by 24% compared to the traditional method,and the fit improved by 19%,which confirms the effectiveness of the method.A 2D distribution detection system for ammonia in farmland was built and the built detection system was used to detect the 2D distribution of static and farmland ammonia concentrations.The 2D distribution detection experiment for static ammonia concentration verifies the effectiveness and stability of the system;the 2D distribution detection experiment for farmland ammonia concentration achieves the purpose of this thesis to reconstruct the ammonia concentration in farmland,and the system can reconstruct the ammonia concentration distribution image well for the test fields in different areas of fertilizer application.