Date of Award

Spring 5-22-2021

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Mechanical and Aerospace Engineering


Jianshun Zhang


Green building design, Heat flow based, Integrated design process, Optimization, Performance evaluation, Simplified and scalable

Subject Categories

Architectural Engineering | Architecture | Engineering | Mechanical Engineering


The overall goal of the study was to develop a “Virtual Design Studio (VDS)”: a software platform for integrated, coordinated and optimized design of green building systems with low energy consumption, high indoor environmental quality (IEQ), and high level of sustainability. My dissertation research was focused on the development of a key VDS component -- an integrated design process and a near-real time performance simulation approach for fast feedbacks at the early stage of the integrated design process.

A design process module “Magic Cube” (MC) was developed for the VDS as the core for the design integration and coordination. It sets up the whole design framework in 3 dimensions: design stages (assess, define, design, apply, and monitor), design factors (site and climate, form and massing, external enclosure, internal configuration, environmental systems, energy systems, water systems, material use and embodied energy, and system interdependencies), and involved teams (architectural design, systems design, and project management). Within this 3D framework, “Task” is proposed as the unit to represent the whole design process with attributes of the design stage, actor, factor, and interdependencies (process and information flow interdependencies). Each individual task is represented in “Input-Process-Output” pattern, composing of: necessary input (quantitative, qualitative, reference, and user-defined information) to support its execution; clearly stated actions need to be performed in order to complete the task; output information will be generated, which be used by next linked tasks (interdependencies). To further facilitate the easy process navigation and management, tasks can also be decomposed and aggregated using the built-in parent and child relationships. Multiple types of views also have been incorporated for better design process visualizations.

Comparing with the traditional design process (often a linear process, teams are involved only when necessary, building systems are designed in isolation with limited optimization among others), the developed MC intended support seamless design transition among stages, enhance multi-disciplinary coordination of (architects, engineers, and project management) team members, and design factors integration through whole building performance analysis.

A Leadership in Energy & Environmental Design (LEED) platinum rated medium-size office building was used as a hypothetical case study to illustrate how the MC method could be applied to achieve a high-performance office building design. From early to the detailed design stage, the building design process and associated design parameters with heavy impacts on the building’s performance were investigated, respectively. The design alternatives with optimal performance were recommended.

As the early stage design decisions have the most and fundamental impact on building performance, a simplified and scalable heat-flow based approach for the form, massing and orientation optimization was developed. A reference building model (RBM) was first defined with pre-selected building materials and heating, ventilation, and air conditioning (HVAC) systems for the intended climate and site conditions. The energy performance of this RBM was estimated by the whole building energy simulation using the detailed EnergyPlus model. Heat fluxes from the enclosure to the indoor air were extracted from the RBM simulation. A simplified physics-based correlation model was developed to predict how these fluxes would be affected by the shape of the building geometry, window-to-wall ratio (WWR), and orientation of a proposed building design. Based on building indoor air space heat balance, the predicted heat fluxes were then used to predict the energy consumption of the proposed building. Compared with conventional detailed energy simulation, this simplified scalable heat-flow prediction method was demonstrated to be 2,500+ times faster (depending on the complexity of the proposed design) with good accuracy. It hence enables effective design evaluations and fast iterations for the early stage HPB design optimization.


Open Access