Design And Implementation Of Remote Monitoring System For Water Quality COD Parameters Based On Ultraviolet Absorption Spectroscopy

The problem of water pollution is still serious,and it is imperative to improve the technical level of water quality monitoring.Based on the UV absorption spectroscopy technique,a linear-exponential segmentation model for the relationship between UV absorbance and COD of water is established by analyzing the UV absorption spectra of potassium hydroxide phthalate with different COD values,the quantitative analysis of a wide range of water COD was realized by the model,and the remote monitoring system base platform of water quality COD parameters was designed.The experimental system of water COD ultraviolet absorption spectrum measurement was constructed with the micro spectrometer as the core,and the UV absorption spectrum of potassium hydrogen phthalate solution was obtained by spectral measurement experiment.The absorption spectra of the samples were studied by analyzing the absorption spectra of different COD samples and their curves with wavelength.the dominant band was selected by the correlation coefficient method combined with the ultraviolet absorption characteristics of the sample,and the dominant wavelength was determined to be 275 nm.Spectral analysis shows that after the sample COD increases to a certain extent,the COD and absorbance of the sample deviate significantly from the linear relationship.An algorithm for determining the segmentation points of the model based on the regression function and dynamic error is designed to improve the fitting accuracy of the segmentation model.The unary robust linear regression and the unary nonlinear least squares regression were selected in low concentration segment and higher concentration segment,respectively,and a linear-exponential segmentation model for the relationship between absorbance and COD of water quality in the range of 40 mg/L to 300 mg/L and 300 mg/L to 560 mg/L was established at the dominant wavelength,which has the highest precision compared with different fitting models,and the prediction effect on the prediction set samples is better.The Raspberry Pi was chosen as the core,DS18B20 is selected as the water temperature sensor,and the GUVB_T21GH ultraviolet sensor is selected to achieve photoelectric conversion and signal amplification,and the MCP3008 chip is used for analog-to-digital conversion.A remote monitoring system for water quality COD parameters was initially developed based on related bus technology,Open-Falcon monitoring framework and IP port forwarding technology,which realized accurate remote monitoring of water COD and temperature after system calibration.The remote monitoring system designed in this paper provides an important foundation and reference basis for further realizing multi-parameter remote real-time online monitoring such as water COD.