Detection Of Humus In Water Based On Three Dimensional Fluorescence Spectrum

Humus widely exists in the actual water body,which has a great impact on water quality.At present,the commonly used research method is to detect humus in water by physical and chemical methods.This method not only takes a long time,but also causes secondary pollution to water body if all kinds of chemicals are not properly treated.At present,in order to solve the above problems,some researchers use fluorescence spectroscopy to study humus in water.Due to the complexity of the experimental data,it is necessary to use a variety of high-dimensional tensor decomposition techniques including parallel factor analysis technology to analyze the data.However,at present,the use process of parallel analysis technology is still unclear,there are many standards for the judgment of the model,and it is difficult for the algorithm to realize the fast analysis of the data.In view of the above problems,this paper designs and builds a three-dimensional fluorescence spectrum detection system applied to water quality analysis.On the basis of learning parallel factor analysis technology,the parallel factor analysis algorithm applied to three-dimensional fluorescence spectrum analysis is compiled by using computer language,which realizes the detection of standard solution and actual water body in the school.The main research contents and achievements are as followsFirstly,the parallel factor analysis algorithm is written in computer language.Aiming at the problems of different evaluation standards and fuzzy model evaluation process,the algorithm is improved.A variety of evaluation methods including core diagonal matrix,core consistent function and constant wavelength residual are used to evaluate the model and clarify the model evaluation process.Aiming at the problem of long computation time,the iterative method is improved to improve the computation speed.After the completion of the above work,starting from the analysis of humus demand,the selection of a variety of equipment including excitation light source,monochromator and spectrometer is carried out,and the three-dimensional fluorescence spectrum detection system is built.The algorithm is tested by 3D fluorescence spectrum of reference materials,and the results show that the decomposition effect of the algorithm is consistent with the standard,and the time is shorter under the condition of changing the iterative method.Secondly,the important components of humus such as fulvic acid and humic acid are analyzed.The two-dimensional spectra of humic acid and fulvic acid were obtained by a spectrometer with excitation wavelength of 405 nm.Meanwhile,the COD values of standard solutions with different concentrations were obtained by national standard rapid digestion spectrophotometry.The relationship between fluorescence intensity and COD was established,and the model of COD detection in water quality by spectrometry was established.The three-dimensional fluorescence spectrum of the standard solution is obtained by using the three-dimensional fluorescence spectrum system,and the factor analysis of fulvic acid solution and humic acid solution is realized by combining with the analysis algorithm.Compared with the previous research results,it is found that the factor decomposition result is more accurate.Finally,the analysis of water body in the campus of Harbin Institute of Technology(Weihai)is realized.The upper water of Dingxiang lake and Riyue Lake were collected for detection,and their three-dimensional fluorescence spectra were obtained.The spectral data at the excitation wavelength of 405 nm and emission wavelength of 520 nm were fitted with the established COD model curve.It showed that the COD of the upper water of the two lakes was between 4-5mg / L,which was consistent with the results of national standard method.The parallel factor analysis algorithm is used to analyze the water body,and the fluorescence index is introduced to explore the impact of seasonal changes on the site water body.