Research On Online Water Quality Monitoring System Of Unmanned Ship Based On Broad Spectrum Technology

With the rapid development of China’s industry and agriculture,the problem of water pollution is becoming more and more serious,which affects people’s quality of life.Using advanced technology to monitor and control the environment is a fundamental way to solve the problem of environmental pollution.As one of the important platforms for grid monitoring of water pollution,unmanned ship has the advantages of fast detection speed,accurate positioning,flexible arrival position and low construction cost.At present,there are two main ways to monitor the water quality of unmanned ships in China.One way is to monitor the water quality by using electrochemical and ultraviolet spectrophotometry sensor detection technology,which has limited monitoring parameters and cannot meet the monitoring requirements.The other way is to use super-large unmanned ship for water quality monitoring,and the detection method mainly adopts reagent method.This way also has some problems,such as high power consumption,secondary pollution,long time and poor endurance.At present,the water quality monitoring of unmanned ship can’t meet the national requirements for water quality monitoring.Wide-spectrum water quality detection technology has the advantages of simple detection process,no need of reagents,multi-parameters,etc.,and has a very broad application prospect in the field of water quality monitoring.This paper designs an unmanned water quality monitoring system based on wide spectrum,which enriches water quality detection parameters and meets the requirements of national water quality monitoring,and is of great significance to the application field of unmanned water quality monitoring.The system mainly includes a water quality monitor and a data transmission unit.using in situ monitoring mode,the monitor is placed in the ship’s equipment well,directly immersed in the liquid to be tested to obtain the absorbance data of a specific band point,and the data is uploaded to the server through the data conversion unit.In the process of unmanned ship operation,the shaking of the hull will greatly introduce noise and stray light interference.In order to reduce its influence,using the Hilbert huang transform（HHT）as the preprocessing method,this method is adaptive and suitable for the nonlinear and unsteady signal de-noising,the results show that the noise reduction effect is good.Spectral scanning experiments of standard solutions with different concentrations with four parameters:chemical oxygen demand（COD）,nitrate nitrogen（NO₃-N）,nitrite nitrogen（NO₂-N）and total phosphorus（TP）.Partial least squares（PLS）,support vector machine（SVM）and random forest（RF）modeling are used respectively.The modeling results are compared,RF is used to model chemical oxygen demand,and the correlation coefficient is R²=0.9991;RF is used to model nitrate nitrogen,R²=0.9959;PLS is used to model nitrite nitrogen,R²=0.996;PLS is used to model total phosphorus,R²=0.9934.Exploring the turbidity,temperature,speed and other factors on the water quality monitoring results of the unmanned ship,and a regressionn model based on the turbidity compensation parameter K,turbidity value and absorbance is proposed to complete the turbidity compensation for chemical oxygen demand and nitrate nitrogen,and establishing a turbidity-absorbance fitting model for five specific band points,turbidity compensation is done for nitrite nitrogen and total phosphorus.Based on the wide spectrum water quality monitoring system,the four parameters of chemical oxygen demand,nitrate nitrogen,nitrite nitrogen and total phosphorus are accurately measured,and the influence of turbidity is compensated,which provides technical support for online water quality monitoring,and is suitable for unmanned ship,station type and buoy type water quality monitoring.