A novel method for total chlorine detection using machine learning with electrode arrays.

Area of Science:

• Analytical Chemistry
• Environmental Science

Background:

• Chlorine is a common water disinfectant, but its reaction with ammonia forms chloramines, interfering with free chlorine measurements.
• Total chlorine serves as a reliable indicator for effective disinfectant levels in water treatment.

Purpose of the Study:

• To develop a novel method for detecting total chlorine in water using an electrode array.
• To establish a robust prediction model for total chlorine concentration based on electrochemical data.

Main Methods:

• Fabrication of a total chlorine sensor utilizing an electrode array.
• Acquisition of cyclic voltammetry curves at varying chloramine concentrations.
• Application of Principal Component Analysis (PCA) and peak sampling for data extraction.
• Development of prediction models using Support Vector Machine (SVM) and Extreme Learning Machine (ELM).

Main Results:

• The study successfully developed and tested a novel sensor for total chlorine detection.
• Support Vector Regression combined with PCA demonstrated the highest predictive accuracy (R² = 0.9689).
• The electrochemical method proved effective in predicting total chlorine concentrations.

Conclusions:

• A simple and effective electrochemical method for determining total chlorine in water has been presented.
• The developed sensor and modeling approach show potential for real-time monitoring of water disinfection and treatment processes.
• This technique offers a viable alternative for assessing disinfectant levels where free chlorine measurements are compromised.