Research On Turbidity Compensation Method Of Water Ultraviolet-visible Spectrum Based On Scattering Spectrum Analysis

With the rapid development of my country’s economy,the environmental problems brought about by the unsustainable and unhealthy development methods in the past have become more and more prominent.As an important part of the country’s “five-in-one”overall layout,the construction of ecological civilization has been elevated to national development.height.The primary task of ecological environment governance is water environment monitoring and governance.In recent years,water pollution caused by substandard discharge of industrial sewage and domestic sewage has caused serious threats to our country’s ecological environment,food safety,and national health.In water environment management and protection,water quality testing is an extremely important link,which is widely used in various fields such as routine management of river sections or drinking water sources,early warning of sudden pollution,and water quality evaluation of sewage treatment plants.In the field of water quality detection,direct ultraviolet-visible spectroscopy detection has the advantages of fast detection speed,low cost,small equipment size,and no secondary pollution.It has become a research hotspot in recent years.Aiming at the problem of turbidity interference caused by light scattering of suspended particles in water quality detection by ultraviolet-visible spectroscopy,the application of common turbidity compensation methods is weak;And the types and contents of organic matter in different measurement scenarios are different,which makes the spectral characteristics of the samples significantly different.A single water quality solution model is only suitable for measurement environments with relatively stable components,which makes it difficult to guarantee the accuracy and reliability of spectroscopic detection.,Hindering the wide application of this method.Based on the above problems,the relevant research work on the multi-scene optimization method of water turbidity compensation and chemical oxygen demand(COD)detection based on scattering spectrum analysis has been carried out.The main contents of the thesis are as follows::1)Introduced the generation mechanism of ultraviolet-visible spectrum,the principle of spectrum analysis based on the absorption characteristics of organic matter and Lambert-Beer law,and the main detection content of ultraviolet-visible light method;introduced the basic theory of Mie light scattering of particulate matter and derives The numerical calculation process of physical quantities related to Mie light scattering is presented..2)Two methods of water body UV-Vis full spectrum correction based on scattering spectrum analysis are studied.Firstly,the scattered light intensity distribution and extinction spectrum of particles of different sizes are analyzed by combining the physical properties of particles with Mie theory.Based on this,a power function modified fitting equation is constructed,and the equation is verified experimentally,and high accuracy is obtained.Achieved accurate spectral calibration;Tried a method of turbidity prediction and turbidity compensation based on multi-angle scattering spectrum acquisition.The experimental results show that this method can measure and calibrate the turbidity introduced by the scattering of suspended particles in the water sample by using the multi-angle scattering spectrum while calculating the content of soluble pollutants in the transmission spectrum.It provides a new reference for the turbidity compensation method for water quality detection by spectroscopy,but its accuracy is not yet ideal,and the optical system needs to be further optimized.3)Researched a multi-scenario optimized spectral classification modeling solution method.First,perform turbidity correction on the sample spectrum through the power function correction fitting equation;use the absorbance normalization method to obtain the linear feature spectra of different scenes to form a scene feature library,and use the Jackard similarity theory to realize the scene matching of the test sample.Finally,the COD concentration calculation is completed based on the partial least squares(PLS)algorithm.This method can greatly improve the COD detection accuracy under multi-scene conditions,effectively reduce the COD calculation error under multi-scene conditions,and help solve the stability and reliability problems encountered in the popularization and application of spectral detection methods.The practical value.