Abstract

Building Performance Simulation (BPS) is frequently used by decision-makers to estimate building energy consumption at the design stage. However, the true potential of BPS remains unrealized if trial and error methods of building simulation are used to identify combinations of parameters to reduce energy use. Optimization techniques combined with BPS offer many benefits such as: (i) identification of potential optimal designs which best achieve desired performance objectives; (ii) system level component integration by simultaneously considering conflicting trade-offs; and (iii) a process-oriented simulation tool that is complementary to BPS, eliminating the need for repetitive userinitiated model evaluations. However, the capability of optimization algorithms to effectively map out the entire solution space and discover information is farther reaching than building design. As shown in this thesis, optimization datasets are also a valuable resource for conducting uncertainty and sensitivity analyses and evaluating policies to incentivize low-energy building design.

Two performance criteria are considered in this thesis: net-energy consumption and life-cycle cost. The term ‘performance-optimized’ refers to the extreme of these two criteria that is Net-Zero Energy (NZE) and cost-optimized buildings. A Net-Zero Energy Building (NZEB) generates at least as much renewable energy on-site as it consumes in a given year. A cost-optimized building has the lowest life-cycle cost over a considered period. A focus of this thesis is identifying optimal pathways to NZE and cost-optimized building designs.

This thesis proposes the following approaches to identify pathways to net-zero energy: (i) a redesign case-study of an existing near-Net-Zero Energy Home (NZEH) archetype using a proposed optimization methodology; (ii) the development of an information-driven hybrid evolutionary algorithm for optimal building design; (iii) a methodology for identifying the influence of design variations on building energy performance; (iv) a methodology to evaluate the effect of incentives on life-cycle energy-cost curves; and (v) effect of a time-of-use feed-in tariff on optimal net-zero energy home design.

The optimization methodology consists of: (i) an energy model; (ii) a cost model; (iii) a custom optimization algorithm; (iv) a database; and (v) a statistics module. Several new simulation techniques are proposed to identify pathways to performanceoptimized net-zero energy buildings: (i) probability distribution functions extracted from previous simulations; (ii) back-tracking searches; and (iii) importance factors to summarize back-tracking search results.

This thesis provides valuable information related to: (i) the development of performancebased energy codes for buildings; (ii) systematic design of cost-optimized NZEHs; (iii) systematic analysis of the impact of different design parameters on energy consumption and cost; (iv) the study of incentive measures for NZEHs.