Dissolved Oxygen (DO) Meter Hydrological Modelling Using Predictive Algorithms

Dissolved oxygen is one of the critical indicators of a body of water's health and water quality. It refers to the presence of free, non-compound oxygen found in water. It also influences the growth and survival of the aquatic organisms living in it. This study aims to develop a low-cost, multi-function device that could determine the value of the dissolved oxygen (DO) level through hydrological modelling of water parameters such as temperature, pH, and conductivity using Decision Tree, Decision forest, and Multi-layer Perceptron machine learning algorithms. Using various metrics, the most efficient model was built using Random Forest algorithm, for it yielded the most reliable metrics when compared to the other two algorithms. The evaluated model has the following metrics: The Coefficient of Determination, or how well a model explains and predicts future outcomes, is 0.99. The Mean Absolute Error, or the average magnitude of the errors in a set of predictions, is 0.32. The Mean Squared Error, utilized in order to measure the performance of an estimator, is 0.36. The Root Mean Squared Error, or how concentrated the data is around the line of best fit, is 0.60. Relative to Atlas Scientific's DO Sensor, the device can predict the dissolved oxygen level of a given water pond with 2.61% error. The final device is a handheld device consisting of the sensors for the highest- ranking parameters with respect to their relationship to DO: temperature, conductivity, and pH.