Electrical energy consumption as a function of urban variables

• Main Author: De Souza, Léa Cristina Lucas
• Publication Date: 2008
• Other Authors: De Oliveira, Alinne Prado
• Format: Conference object
• Language: eng
• Source: Repositório Institucional da UNESP
• Download full: http://hdl.handle.net/11449/231435
• topic: Artificial neural networks; Energy consumption; Sky view factors

This paper analyzes the electrical energy consumption of households as a function of urban variables, by modelling the urban thermal environment with Artificial Neural Networks (ANN). The study area was a residential neighbourhood. Urban features of reference points were determined by the following characteristics: urban heat island, sky view factor, and users' income level. For each of these reference points, urban air temperatures at the pedestrian level were collected with data-loggers. At the same time, rural temperatures made available by the city meteorological station site were registered. In addition, the user's profiles were identified by means of a questionnaire applied to the households. Their electrical energy consumption data were also collected from the power supply company. Models applying Artificial Neural Networks were then developed for the most important periods of UHI intensity. The results show that low values of sky view factor and high urban heat islands, when observed in high income zones, are associated with the largest electrical energy consumption patterns.