Development of a decision-support model based on fuzzy logic for optimizing of High Energy Performance (HPE) housing design in Algeria.

Abstract

In Algeria, the final energy consumption remains dominated by the housing sector, which represents 36% of the total final consumption. In 2019, the energy use increased more than 80 % from 2009. Thus, the prospects for developing the housing stock will lead to an exponential increase in this energy consumption. However, Buildings in the residential sector nevertheless have a significant potential for energy savings. In this context, the construction of energy-efficient housing is essential for controlling energy consumption in the residential sector. The performance optimization of low-energy buildings should always take place in the early design stages when most of the critical decisions affecting building energy performance are made by integrating the optimal values of different building parameters depending on the climatic conditions. To design and construct low-energy buildings, it is essential to assure informed decision-making during the early design phases. Therefore, there is a need for the development of decision support tools that can predict the building performance and support the design decision making of low-energy buildings. This research aims to contribute to the implementation of energy-efficient housing buildings across the Algerian territory and under all Algerian climate zones through informed design decision making in the early design stages of low-energy building. Therefore, this thesis developed a decision support model that could estimate building energy performance (cooling and heating energy loads) in early design stages without using building performance simulation tools. The model provides rapid, energy-relevant feedback, and visualize possible consequences of the design decisions. Initially, the bioclimatic potential of all Algerian climate zones has been investigated using a dual approach that combines psychrometric chart-based analysis with building performance simulation analysis (EnergyPlus) to provide accurate bioclimatic design recommendations. Afterwards, the thermal and energetic behaviour of the typical multi-family apartment buildings, across the Algerian territory, has been evaluated using BPS techniques (EnergyPlus) combined with GIS to generate a new spatial distribution map for energy demand and thermal comfort estimation in Algeria. These maps will inform building designers without accessing, analyzing, or interpreting dense textual information. Then, the typical multi-family apartment building design has been optimized for each climate zone using a mixed approach that combine between building performance simulation (BPS) tool (EnergyPlus) and building performance optimization (BPO) algorithm (NSGA-II). Finally, this research ends by developing a design decision-making model based on prediction using an Adaptive Neuro-Fuzzy Inference System (ANFIS) to estimate the Abstract II cooling and the heating energy loads of the typical multifamily social residential building through the building design parameters variation. As a result, this thesis provided design recommendations for each climate zone and a decision-making model that presented a high accuracy level.