Possible cost saving potentials in planning and construction of high performing nearly zero energy buildings (nZEBs) with advanced energy standards are often not sufficiently assessed, as only a few, out of numerous possible variants of technology sets are considered in the traditional planning process. Often planning and analysis are not carried out in parallel, and the alternative technical options are discarded at an early stage. If, on the other hand, possible variants are realistically compared in the planning phase, a profound decision can be made. nZEB-design is also a multi-objective optimization problem where stakeholder interests' conflict with each other. This research addresses a methodological approach to better understand the effects that technical variables have on energy, environmental and economic performance over the whole life cycle of a multi-family residential building in Sweden. The research goal is to identify the most significant technical nZEB design variables organized into a consistent framework. In this paper, in a first step an exhaustive search method is assessed for a multi-family residential building in Sweden that systematically investigates all possible variant combinations. In a second step the derived results are applied to multiple objectives and optimisation goals for a multi-target decision-making framework so that different actors can decide between optimal solutions for different objectives. This approach seeks to explore a set of optimal solutions rather than to find a single optimal solution. On the one hand, a variety of technologies, such as insulation of the building envelope, ventilation or electricity and heat supply, and on the other hand a variation of the boundary conditions (such as observation period, user behaviour, energy price increases or CO2 costs) was investigated. The results were analysed energetically and economically over the life cycle of the building with the objectives of identifying coherences, deriving trends and optimizations over a time span of 40 years. The results show that the variance in the financing costs (20%) and the net present value (15%) is relatively low, whereas the primary energy demand (66%) and the CO2 (73%) emission vary in a broader range. The optimum cost curve in relation to CO2 emissions is very flat. Low emissions and energy requirements can, therefore, be achieved with different energy concepts as long as the envelope is very efficient. Due to the nature of an exhaustive search approach, it is also possible to find technical solution sets and design strategies with nearly equal financing cost and/or net present values, but with less primary energy consumption and/or CO2 emissions.
ExhauFS: exhaustive search-based feature selection for classification and survival regression
