| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 6.07 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O presente trabalho tem como principal objetivo a aplicação do método de Monte Carlo ao
estudo da influência dos dados de entrada na simulação energética de edifícios escolares.
Pretende-se perceber como se comporta a energia das salas de aula perante a variabilidade dos
parâmetros de entrada. Para tal, foram definidos alguns dados de entrada da simulação como
variáveis aleatórias com uma determinada média e desvio padrão e com distribuição normal.
As variáveis escolhidas para este caso de estudo foram a densidade de pessoas, o
metabolismo, a ventilação (considerando dois cenários distintos), a iluminação e a
condutibilidade térmica do reboco interior e da cobertura.
Os valores das amostras das variáveis de estudo foram obtidos aleatoriamente, através do
método de Monte Carlo. Este método consiste em gerar valores aleatórios, assumindo uma
determinada distribuição de probabilidade, criando vários cenários para essas variáveis. De
forma a melhorar a eficiência do procedimento foi utilizado o método Hipercubo Latino.
Foram geradas amostras de 25, 50, 100, 200 e 500 valores aleatórios para todas as varáveis de
estudo com o objetivo de permitir efetuar uma análise de sensibilidade relativamente à
exatidão e, consequentemente, ao número de casos necessários em estudos desta natureza.
Depois das amostras serem geradas foram criados os respetivos ficheiros de dados e efetuadas
as simulações utilizando o programa de simulação térmica e energética EnergyPlus. Todo o
procedimento foi automatizado com recurso a macros programadas em Excel.
Os resultados obtidos nas várias simulações foram analisados estatisticamente e tratados
segundo uma análise descritiva. Foi efetuada uma análise às necessidades energéticas das
salas de aula numa base mensal e relativamente ao total anual. Verificou-se que, dos
parâmetros estudados, o que mais influencia as necessidades energéticas de aquecimento das
salas de aula é a ventilação, concluindo-se que existe uma forte correlação linear entre ambos
os parâmetros.
ABSTRACT: The present work has as main objective, the application of the Monte Carlo method in the study of the influence of the entrance data on the energetic simulation of scholar buildings. We intent to understand how the classroom energy behaves before the variability of the entrance parameters. For this, some entrance data simulation were defined as random variables with a certain average and pattern deviation and with normal distribution. The variables chosen for this case study were the density of people, the metabolism, the ventilation (considering two distinct scenarios), the illumination and the thermic conductibility of the interior plaster and the roof covering. The values of the samples variables of the study were obtained randomly, through the Monte Carlo method. This method consists of generating random values, assuming a certain probabilities distribution and in creating several scenarios for those variables. To improve the efficiency of this method it was used the Hipercubo Latino method. Samples were generated of 25, 50, 100, 200 and 500 random values for all the study variables with the main goal of allowing performing a sensibility analysis, regarding the accuracy and, consequently, the number of necessary cases in studies of this nature. After the samples were generated, the respective data files were created and the simulations were done using the thermic and energetic simulation EnergyPlus program. All the procedure was automatized using the macros programmed in Excel. The results obtained in the several simulations performed were statistically analysed and treated according a descriptive analysis. It was done an analysis to the monthly and total annual energetic needs of the classrooms. It was found that from the parameters studied, the one that most influences the energetic needs of heating in a classroom are the ventilation, concluding that there’s a strong linear correlation between both parameters.
ABSTRACT: The present work has as main objective, the application of the Monte Carlo method in the study of the influence of the entrance data on the energetic simulation of scholar buildings. We intent to understand how the classroom energy behaves before the variability of the entrance parameters. For this, some entrance data simulation were defined as random variables with a certain average and pattern deviation and with normal distribution. The variables chosen for this case study were the density of people, the metabolism, the ventilation (considering two distinct scenarios), the illumination and the thermic conductibility of the interior plaster and the roof covering. The values of the samples variables of the study were obtained randomly, through the Monte Carlo method. This method consists of generating random values, assuming a certain probabilities distribution and in creating several scenarios for those variables. To improve the efficiency of this method it was used the Hipercubo Latino method. Samples were generated of 25, 50, 100, 200 and 500 random values for all the study variables with the main goal of allowing performing a sensibility analysis, regarding the accuracy and, consequently, the number of necessary cases in studies of this nature. After the samples were generated, the respective data files were created and the simulations were done using the thermic and energetic simulation EnergyPlus program. All the procedure was automatized using the macros programmed in Excel. The results obtained in the several simulations performed were statistically analysed and treated according a descriptive analysis. It was done an analysis to the monthly and total annual energetic needs of the classrooms. It was found that from the parameters studied, the one that most influences the energetic needs of heating in a classroom are the ventilation, concluding that there’s a strong linear correlation between both parameters.
Description
Mestrado em Engenharia de Construção e Reabilitação
Keywords
Edifícios escolares Simulação energética Método Monte Carlo Análise estatística
Citation
Publisher
Instituto Politécnico de Viseu. Escola Superior de Tecnologia e Gestão de Viseu