Study On The Algorithm Of COD Monitoring Sensor Based On The Partial Least-square Regression And Turbidity Compensation

With the increasing pollution of water resources,the government efforts to strengthen the work of water quality monitoring,a water quality monitoring equipment with fast,accurate and on-line monitoring function is the key to improve the level of monitoring and enforcement efficiency of law.Chemical oxygen demand(COD)is an important index to reflect the content of organic pollutants in water,it is one of the comprehensive indexes to evaluate the degree of water pollution,so the senor which monitoring the COD is an important instrument.At present,there are many kinds of water quality monitoring sensors of COD at home and on abroad,but the main methods of detecting COD can be roughly divided into two categories: the first category is based on chemical analysis methods to detect COD;the second category is detected by physical methods.The chemical method is mainly represented by potassium permanganate method and potassium permanganate method,it is characterized by high measuring precision,long time consuming,complicated pretreatment process of water sample,the use of chemical reagents prone to secondary pollution.The physical method is mainly represented by the ultraviolet absorption method(UV method),it is characterized by high sensitivity analysis,no need for pretreatment of water samples,no secondary pollution,easy operation,fast analysis speed.Nowadays,there is a sudden water pollution accident.and the monitoring of water quality can not be interrupted,so the COD monitoring sensor for water quality designed by chemical methods can not meet the above requirements.At the same time,there are many defects in the design of COD monitoring sensor based on the principle of UV absorption.Therefore,the purpose of this paper is to improve the spectrum of COD monitoring sensor based on the principle of UV absorption,and make the COD spectrum monitoring sensor toimprove the measurement accuracy and extend its application scope,in order to produce an water quality sensor which can be continuously monitored on-line.This paper first systematically introduces the basic principle of UV absorption,and the development and advantages and disadvantages of several commonly used methods in UV absorption,the basic principles and related properties of partial least squares regression are introduced in detail.This paper analyzes the results of previous studies,in previous studies,many scholars used the UV absorption of the liquid to be measured at a specific wavelength to predict the COD value of the water sample;on the other hand,some scholars have considered the influence of turbidity on the absorption at specific wavelengths,and made the turbidity compensation for the specific wavelength;and some scholars are considering using the measured liquid at several specified wavelength absorption to predict the COD value of water samples.However,no scholars have considered the absorption of liquid to be measured at several specified wavelength to build the model,At the same time,considering the influence of turbidity.In order to improve the accuracy and scope of COD monitoring.This paper presents a COD detection algorithm based on partial least squares regression and turbidity compensation,by using algorithm which combining ultraviolet absorption spectroscopy with partial least squares regression to predict the chemical oxygen demand in water.At the same time,the influence of turbidity on the absorption of modeling was considered,and the effect on turbidity was compensated.The experimental results show that the proposed method is suitable for different types of sewage,the average relative error is less than5%,and the prediction accuracy is better than the partial least squares regression model without turbidity compensation.It provides a basis for the development of an adaptive and on-line monitoring of water quality COD sensor.