Multi-objective optimization for school buildings retrofit combining artificial...

• Autor(es): Almeida, Ricardo ; Freitas, Vasco
• Data: 2013
• Identificador Persistente: http://hdl.handle.net/10400.19/1953
• Origem: Repositório Científico do Instituto Politécnico de Viseu
• Assunto(s): Multi-objective optimization; school building; artificial neural networks; life cycle cost
• Tipo de Documento: Artigo
• Idioma: Inglês

Descrição

The renovation of a school building should be regarded as a process of combining a number of variables and objectives, sometimes conflicting, including energy, indoor environmental quality and costs (initial, operational and maintenance), on a search for an "optimum solution". This multi-objective optimization procedure is particularly important in a time of severe economic crisis, with few available financial resources and, as such, their management and the investment decisions require great prudence from the decision maker. In this research a methodology to optimize the insulation thickness of external walls and roof, in the retrofit of two school buildings, is proposed. The school performance was defined considering two objectives: the annual heating load and the discomfort in the classrooms due to overheating. The calculation of the performance functions implies an annual simulation of the building and Artificial Neural Networks were training to approximate them. The minimization of the Life Cycle Cost of external walls and roof retrofit allowed the economic optimization of the insulation width.