Browsing by Author "Gavilánez Barrionuevo, Christian Sebastián"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Estimación de la calidad del aire en ambientes interiores en laboratorios químicos de la Universidad Técnica de Ambato utilizando redes neuronales artificiales
    (Universidad Técnica de Ambato. Facultad de Ciencia e Ingeniería en Alimentos y Biotecnología. Carrera de Ingeniería Bioquímica, 2021-01) Gavilánez Barrionuevo, Christian Sebastián; Córdova Suárez, Manolo Alexander
    According to the US Environmental Protection Agency, the levels of contamination in closed environments can be 10 to 100 times higher than outdoors. The objective of this research project was to estimate the air quality in indoor environments of chemical laboratories. from the Technical University of Ambato through the use of Artificial Neural Networks. A Backpropagation RNA model was used, an algorithm capable of organizing in the intermediate layer or layers, between the input and the output of the network. Measurements were made at 25 laboratories with high carbon dioxide output rates, an MX6 iBRID equipment was used for direct measurement of carbon monoxide and carbon dioxide, the Extech HT30 for temperature and humidity measurement and for measurement of particulate material the Aerocet 831 portable particle counter. The sampling points were chosen in accordance with the UNE 171330-2: 2014 standard. A cascade network with 2 hidden layers was configured with a gradient descent training, with 7 epochs, resulting in a performance of 80 percent. Having a linear trend line and a value of R = 0.99557 with which it can be affirmed that the 7 epochs of training is the optimal number to avoid failures in the network. This model can be adapted for the forecast of other air pollutants, whether chemical or particulate matter, with good precision and in a simple way.