Maestría en Obras Hidráulicas

Permanent URI for this collectionhttps://repositorio.uta.edu.ec/handle/123456789/44309

Browse

Filters

20251

Settings

Author: search.filters.author.Llamuca Montaluisa Segundo Atahualpa

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Generación de un modelo estadístico para la automatización del control de los parámetros de calidad del agua, de la planta de tratamiento de agua potable El carrizal, perteneciente a la parroquia San Miguel, del Cantón Salcedo, Provincia de Cotopaxi
    (Ingeniería Civil y Mecánica. Carrera Ingeniería Civil. Maestría con mención en Obras Hidráulicas, 2025-12-15) Llamuca Montaluisa Segundo Atahualpa; Bayas Altamirano Myriam Marisol
    The current research aims to generate a statistical predictive model based on machine learning methods, geared towards automating the control of water quality parameters at the "El Carrizal" treatment plant, located in the Salcedo canton, Cotopaxi province. To achieve this, a mixed-methods approach with a quantitative focus was employed, broken down into four stages: data preprocessing, predictive model design, benchmarking, and operational validation. The analysis of pH, turbidity, and residual chlorine parameters utilizes historical data from IoT sensors and applies multiple linear regression algorithms, support vector machines, decision trees, multilayer neural networks, and random forests. The results show a low linear correlation between the variables (R² values close to zero), which can be explained by the system's operational stability and the nonlinear complexity of the hydraulic system. Despite this, the Random Forest model stood out for its good balance between mean error, stability, and generalizability, thus serving as a base model for predictive process automation. These findings propose a replicable methodology based on data and machine learning, paving the way for more efficient and sustainable intelligent systems for water quality control.