Optimizing architectural and structural aspects of buildings towards higher energy efficiency

Álvaro Fialho, Youssef Hamadi, Marc SchoenauerGECCO (Companion) 2011 · DOI

TL;DR

This paper presents a framework for automatically optimizing building designs to simultaneously minimize energy consumption and construction costs. By combining a sophisticated evolutionary algorithm (Hype) with an energy simulation tool (EnergyPlus), the authors demonstrate a method to explore design trade-offs and discover efficient, cost-effective building configurations.

What Problem Does This Paper Address?

The escalating global energy consumption, largely driven by the building sector, presents a significant challenge for sustainable development. Buildings are responsible for a substantial portion of energy use and associated carbon emissions. While optimizing for energy efficiency is crucial, it often leads to increased construction costs. Furthermore, design decisions involve complex trade-offs, such as balancing natural daylighting with thermal insulation, or minimizing energy use without compromising occupant comfort. This paper tackles the problem of finding optimal building designs that navigate these conflicting objectives, aiming to reduce both environmental impact and economic expenditure.

Methods

The core of the proposed approach is a multi-objective optimization framework that leverages two key components:

  • EnergyPlus: A widely-used, free energy analysis and thermal load simulation program. It models building performance based on user-defined design parameters and local meteorological data, providing a quantitative measure of energy consumption.
  • Hype: A state-of-the-art multi-objective evolutionary algorithm specifically designed for optimizing problems with many objectives. Hype employs a novel fitness assignment and replacement mechanism based on the hyper-volume measure, which quantifies the volume of the objective space dominated by a set of solutions.

The workflow involves the Hype algorithm exploring a design search space, generating candidate building configurations. These configurations are then evaluated by EnergyPlus for their energy consumption. Hype uses this feedback to iteratively refine its search, guiding the population of solutions towards a Pareto front—a set of non-dominated solutions where no objective can be improved without degrading another.

The paper also details the implementation of Hype, including its mixed representation (real values for orientation, integers for material choices), tournament selection, and simple variation operators (crossover and Gaussian mutation).

Datasets

The paper uses a specific commercial building design from an existing exercise (EnergyPlus documentation) as its primary test case. This building is described with its dimensions, zone layout, window specifications, material properties for walls and roof, and HVAC system details. The environmental data used for simulation is from Chicago, IL, USA (TMY2 data).

A custom database of 33 insulation materials is created, each with an associated cost (€/m²) and thermal resistance (m² K/W). This database is derived from a French retailer's pricing.

Key Insights and Ideas

  • Automated Design Exploration: The framework automates the complex and time-consuming process of exploring architectural and structural design options, offering architects a powerful tool to discover efficient solutions.
  • Navigating Conflicting Objectives: The paper highlights the inherent trade-offs in building design (e.g., energy efficiency vs. cost) and demonstrates how multi-objective optimization can effectively find compromises that satisfy multiple criteria simultaneously.
  • The Power of Evolutionary Algorithms: The use of Hype showcases the capability of evolutionary algorithms to efficiently search large and complex design spaces, even with multiple, competing objectives.
  • Empirical Validation of Design Principles: The experiments provide empirical evidence for how specific design choices, like building orientation and insulation material selection, impact energy consumption and cost.

Key Results

The preliminary experiments focused on optimizing a single-floor commercial building by varying its orientation angle and the materials used for two insulation layers in its exterior walls. The objectives were to minimize total yearly energy consumption (kWh) and construction costs (€/m²).

  • The original building design (orientation 0°, specific materials) resulted in an annual energy consumption of 65735.2 kWh.
  • The optimized solutions achieved a reduction of around 5% in energy consumption compared to the original design.
  • The study identified 21 non-dominated solutions, representing a Pareto front of optimal trade-offs between energy consumption and cost.
  • For instance, one solution offered the most energy-efficient outcome with a ~5% reduction, while another provided a significant cost saving while still achieving a notable reduction in energy use.
  • Optimizing the building's orientation angle alone (for a fixed set of materials) showed that a specific angle (around 265 degrees) could save up to 700 kWh (approximately 1%) of energy consumption.

Limitations and Future Work

The authors acknowledge that this work is in its early stages and has several limitations:

  • Simplified Scope: The current experiments consider only a single, simple building and focus on a limited set of design variables (orientation and insulation materials).
  • Operational Energy Focus: The optimization primarily considers energy consumption during the building's operational phase, excluding embodied energy from construction and demolition.
  • Passive Design Emphasis: The current objectives are limited to passive energy reduction, not actively incorporating HVAC system design or control strategies.
  • Limited Parameter Tuning: While the Hype algorithm is state-of-the-art, the specific variation operators and parameters used were not extensively tuned for this particular problem.

Future work proposed includes:

  • Expanding Design Variables: Incorporating more complex design aspects like window size and placement, and considering natural daylighting as an additional objective.
  • Including HVAC Systems: Optimizing the design and control parameters of HVAC systems.
  • Broader Building Typologies: Testing the framework on a diverse range of building types (offices, warehouses, schools, etc.) and geometries to create a benchmark set.
  • Considering Embodied Energy: Incorporating the energy costs associated with construction materials and processes.

Summary updated

BibTeX
@inproceedings{fialho2011optimizing,
  author = {Álvaro Fialho and Youssef Hamadi and Marc Schoenauer},
  title = {{Optimizing architectural and structural aspects of buildings towards higher energy efficiency}},
  booktitle = {GECCO (Companion)},
  year = {2011},
  doi = {10.1145/2001858.2002077},
}
Figure