International Journal on Advanced Science, Engineering and Information Technology, Vol. 11 (2021) No. 6, pages: 2369-2374, DOI:10.18517/ijaseit.11.6.13598

Water Quality Prediction and Detection of the Vibrio Cholerae Bacteria

Camilo Enrique Rocha Calderón, Octavio José Salcedo Parra, Sebastián Camilo Vanegas Ayala


This document shows the results for two water quality-related trials based on the Physico-chemical characteristics given by the used dataset; both trials were carried out based on the same dataset from which the membership sets, and functions were defined the most relevant features. The first trial was a neural network method aimed to predict water quality through attributes as the pH, temperature, turbidity, salinity, among others; the second trial was a fuzzy logic system method for the detection of the Vibrio Cholerae in the water through the usual variables associated to its presence: temperature, salinity, phosphates, and nitrites' levels. The method for this research is divided into two phases. The first phase is developing suitable software using an iterative and incremental process model based on prototypes. The second phase or operative phase has an experimental characterization that allows for an adequation of the environment to establish the main features and properties that are relevant to the study object. The results showed effectiveness values of 99.99% (highest obtained value) for trial one and 70.23% for trial two; such values depict an accurate prediction on the quality of water and a valuable detection for Cholera related bacteria in water supplies. This research developed two highly interpretable and transparent systems to people through the graphic of the correspondences between the rules established and the membership functions in the input and output sets.


Fuzzy systems; neural networks; quality; Vibrio Cholerae; water.

Viewed: 1179 times (since abstract online)

cite this paper     download