Análisis de emisiones de ruido por el uso de generadores eléctricos en la crisis energética de 2024 en el cantón Cuenca

Abstract

The 2024 energy crisis in Ecuador, triggered by severe drought, significantly reduced hydroelectric generation and led to prolonged blackouts. In response, the widespread use of electric generators raised concerns about environmental and health impacts. This study proposes a methodology to estimate pollutant emissions (CO₂, CO, HC, NOₓ) based solely on noise levels. A total of 248 generators were measured, and machine learning models were trained. The decision tree model showed the best performance (R² of 0.66 for CO₂). It was then applied to estimate emissions from 710 generators over 89 days. Results revealed substantial pollutant output. Moreover, noise levels exceeded regulatory limits, affecting public health.