A Novel Determination of Chemical Oxygen Demand (COD) Using Thermal Sensor: A Key Initiative in Wastewater Treatment

Water pollution is a pressing global concern, and the methodologies employed for water assessment play a pivotal role in gauging its severity. The quantification of organic compounds within water bodies assumes paramount significance in the meticulous monitoring and management of pollution levels. India, a nation grappling with the deteriorating health of its rivers, faces an enduring predicament. A staggering 40 million liters of wastewater are introduced into rivers and other aquatic reservoirs every day, with a mere 37% of it undergoing appropriate treatment. At the outset, Chemical oxygen demand (COD) is widely used as one of the most important an index for the appraisal of water quality. The paper describes the design and development of a novel portable Chemical oxygen demand (COD) detection method based on thermal sensors. The shortage of sustainable water resources and increasing awareness of environmental protection refers to the desideratum for an onsite, environmentally friendly, and rapid analytical method. The proposed work utilizes the principle of change in degree of enthalpy i.e., heat generated during the oxidation of organic compounds, and the heat is proportional to the amount of organic compounds contained in the water sample is a part of the thermal sensor’s development. The presented work utilizes Sodium hypochlorite (NaClO) solution is employed as an oxidizer after examining several other oxidizers. Introduced method results to claim, further comprising with COD and were compared against the conventional method using Potassium dichromate. The COD values of water samples from various sources were correlated with those obtained using the standard dichromate method, yielding a linear regression equation of CODts = 1.0215 CODCr − 27.725 and a performing correlation coefficient of 0.9838. The paper compares the results of a few best-performing Machine learning (ML) algorithms like Linear regression, Regression tree, Random forest, and decision tree with grid search. At the outset, the most dynamic model was observed to be the Decision tree with grid search achieving an overall accuracy i.e., the determination coefficient (R2) being 97.85%. Furthermore, It’s important to establish a new environmentally friendly chemical oxygen demand detection method with safe and high efficiency in the future.